Determination and Implications of Electroosmotic Transport In

DETERMINATION AND IMPLICATIONS OF ELECTROOSMOTIC TRANSPORT IN

THE BRAIN

Yifat Guy

BS, University of Pittsburgh, 2004

Submitted to the Graduate Faculty of

Arts and Sciences in partial fulfillment

of the requirements for the degree of

Doctor of Philosophy

University of Pittsburgh

2011

UNIVERSITY OF PITTSBURGH

Graduate Arts and Sciences

This dissertation was presented

Yifat Guy

It was defended on

December 15, 2010

and approved by

Dr. Shigeru Amemiya, Professor, Department of Chemistry

Dr. Adrian Michael, Professor, Department of Chemistry

Dr. Mats Sandberg, Professor, Department of Anatomy and Cell Biology, Gothenberg

University

Dissertation Advisor: Dr. Stephen G. Weber, Professor, Department of Chemistry

2011

Reproduced with permission from:

Guy Y, Sandberg M, Weber SG. Biophysical Journal. 2008;94(11):4561-4569.

Guy Y, Muha RJ, Sandberg M, Weber SG. Analytical Chemistry. 2009;81(8):3001-3007.

iii DETERMINATION AND IMPLICATIONS OF ELECTROOSMOTIC TRANSPORT

IN THE BRAIN

Yifat Guy, PhD

University of Pittsburgh, 2011

Electroosmotic flow is a bulk fluid flow that influences solute transport through capillary

conduits and porous material under the influence of an electric field. The magnitude of

electroosmotic flow is proportional to the ζ-potential of the capillary or porous material. A

porous material such as living tissue would need to have an appreciable ζ-potential to create electroosmotic flow in the interstitial space. Transporting solutes in and out of tissue by virtue of electroosmotic flow in principle has advantages such as avoiding pressure effects and sample dilution that accompany micropush-pull and microdialysis approaches. In order to assess the viability of this approach, it is necessary to know the ζ-potential in the tissue of interest. To address this, a method and apparatus was developed to measure the ζ-potential and tortuosity in tissue slices. The method was applied to organotypic hippocampal slice cultures. The apparatus was improved on in order to provide feedback control to maintain a constant electric field through the tissue culture. The ζ-potential of the organotypic tissue culture is -22 ± 0.8 mV and the tortuosity is 2.24 ± 0.10. With a ζ-potential of -22 mV, low electric fields applied to the brain will create electroosmotic flow.

Electroosmotic flow can be directed to transport extracellular fluid from brain tissue into a conduit such as a sampling capillary. Furthermore, electroosmotic flow can be used in the opposite way to eject fluid and solutes from a capillary or pipette into brain tissue. The

iv electroosmotic effect may be important in the widely used solute delivery method of iontophoresis. In iontophoresis, solutes are ejected into tissue via an applied current. Once in the tissue, solute transport is affected by the electroosmotic flow in the tissue and thus depends on the tissue ζ-potential. The dependence of solute transport on ζ-potential is illustrated using a set of poly(acrylamide-co-acrylic acid) hydrogels. A method of measuring the thickness of organotypic tissue cultures has also been developed. Characterization of ζ-potential and tortuosity provides the fundamentals for understanding electroosmotic flow through the extracellular space of brain tissue.

v TABLE OF CONTENTS

PREFACE ...... XIV

1.0 INTRODUCTION ...... 1

1.1 ELECTROKINETIC TRANSPORT ...... 2

1.2 PHYSICOCHEMICAL BRAIN PROPERTIES ...... 4

1.3 THE MODEL SYSTEM: ORGANOTYPIC HIPPOCAMPAL SLICE

CULTURE ...... 5

1.4 QUANTITATIVE ANALYSIS OF THE EXTRACELLULAR SPACE ...... 7

2.0 INITIAL MOTIVATION ...... 9

3.0 DETERMINATION OF ZETA POTENTIAL IN RAT ORGANOTYPIC

HIPPOCAMPAL CULTURES...... 15

3.1 INTRODUCTION ...... 16

3.1.1 Theory ...... 19

3.2 MATERIALS AND METHODS ...... 21

3.2.1 Chemicals and solutions ...... 21

3.2.2 Instrumentation ...... 22

3.2.2.1 Capillary electrophoresis: ...... 22

3.2.2.2 Apparatus for tissue experiments ...... 23

3.2.2.3 Ag/AgCl electrodes ...... 23

vi 3.2.2.4 Injection capillary ...... 24

3.2.2.5 Imaging ...... 24

3.2.3 Procedures ...... 25

3.2.3.1 Hippocampal organotypic slice cultures ...... 25

3.2.3.2 ζ-potential determination ...... 26

3.2.3.3 Two-fluorophore experiments ...... 26

3.3 RESULTS ...... 26

3.4 DISCUSSION ...... 31

3.4.1 Analysis of the method ...... 31

3.4.2 Analysis of the results ...... 33

3.4.3 Potential applications and implications ...... 36

3.5 ACKNOWLEDGMENTS ...... 38

3.6 SUPPORTING MATERIAL ...... 38

3.6.1 Results ...... 38

3.6.2 Discussion ...... 39

3.6.2.1 Analysis of the method ...... 39

3.6.3 Materials and methods ...... 39

3.6.3.1 Chemicals and solutions ...... 39

3.6.3.2 Capillary electrophoresis ...... 39

4.0 DETERMINATION OF ζ-POTENTIAL AND TORTUOSITY IN RAT

ORGANOTYPIC HIPPOCAMPAL CULTURES FROM ELECTROOSMOTIC

VELOCITY MEASUREMENTS UNDER FEEDBACK CONTROL ...... 41

4.1 INTRODUCTION ...... 42

vii 4.2 THEORY ...... 44

4.3 EXPERIMENTAL SECTION ...... 46

4.3.1 Chemicals and solutions ...... 46

4.3.2 Organotypic hippocampal slice cultures ...... 47

4.3.3 Four-electrode potentiostat ...... 47

4.3.4 Apparatus for tissue experiments ...... 49

4.3.5 Injection capillary ...... 50

4.3.6 Imaging ...... 51

4.3.7 ζ-potential determination ...... 51

4.4 RESULTS ...... 53

4.5 DISCUSSION ...... 57

4.5.1 Comparison of methods ...... 57

4.5.2 Accuracy of the method ...... 60

4.6 SUPPORTING INFORMATION ...... 63

5.0 IONTOPHORESIS INTO A POROUS MEDIUM DEPENDS ON THE ZETA

POTENTIAL OF THE MEDIUM ...... 65

5.1 INTRODUCTION ...... 66

5.1.1 Theory ...... 68

5.2 EXPERIMENTAL ...... 71

5.2.1 Chemicals and Solutions ...... 71

5.2.2 Fluorophores ...... 72

5.2.3 Determination of Diffusion Coefficients ...... 72

viii 5.2.4 Electrophoretic Mobility Determination by Capillary Zone

Electrophoresis (CZE) ...... 73

5.2.5 Synthesis of Hydrogels ...... 74

5.2.6 ζ-Potential and Tortuosity ...... 75

5.2.7 Porosity of Hydrogels ...... 76

5.2.8 Conductivity of Hydrogels ...... 77

5.2.9 Organotypic Hippocampal Slice Cultures (OHSCs) ...... 77

5.2.10 Electrokinetic Ejections into Hydrogels ...... 78

5.2.11 Electrokinetic Ejections into OHSCs ...... 80

5.2.12 Two-pipette experiments ...... 81

5.3 RESULTS ...... 82

5.3.1 Properties of the fluorophores, hydrogels, and iontophoresis capillaries 82

5.3.2 Two-pipette experiment ...... 84

5.3.3 Time to steady-state ...... 85

5.3.4 Correlation between ζ-potential and solute penetration distance: ...... 86

5.3.5 Effect of current on concentration and Pe ...... 88

5.3.6 Comparison of PeE with PeT ...... 89

5.3.7 Small, highly charged solute: ...... 90

5.4 DISCUSSION ...... 91

5.4.1 Fluorophore, hydrogel, and pipette parameters ...... 91

5.4.2 Time to steady-state ...... 92

5.4.3 Solute penetration distance ...... 93

5.4.4 Two-pipette experiments ...... 94

ix 5.4.5 PeE versus PeT ...... 95

5.4.6 Ejections into OHSCs ...... 98

5.4.7 Conclusion ...... 99

5.4.8 Acknowledgement ...... 99

5.5 SUPPORTING INFORMATION ...... 100

5.5.1 Comsol Simulations:¥ ...... 102

6.0 A SIMPLE METHOD FOR MEASURING ORGANOTYPIC TISSUE SLICE

CULTURE THICKNESS ...... 110

6.1 INTRODUCTION ...... 111

6.2 MATERIALS AND METHODS ...... 112

6.2.1 Chemicals and solutions ...... 112

6.2.2 Tissue culturing procedure ...... 113

6.2.3 Tissue thickness measurements ...... 114

6.2.4 Viability measurements ...... 115

6.2.5 Method validation ...... 116

6.3 RESULTS ...... 117

6.3.1 Tissue culture thickness ...... 117

6.3.2 Viabilty ...... 120

6.3.3 Method validation ...... 120

6.4 DISCUSSION ...... 120

REFERENCES ...... 123

x LIST OF TABLES

Table 3.1. Values of µep in HBSS ...... 22

Table 3.2. A summary of the average observed mobilities in tissue and their SEMs ...... 30

Table 3.3. Dye concentrations ...... 40

Table 3.4. CE parameters ...... 40

Table 4.1a. Comparison of observed mobilities ...... 55

Table 4.1b. Comparison of variances ...... 55

Table 4.2a. Comparison of ζ-potential and tortuosity of the two experimental methods ...... 56

Table 4.2b. Comparison of variances ...... 56

Table 4.3a. Comparison of ζ-potential and tortuosity of the two analysis methods ...... 57

Table 4.3b. Comparison of variances ...... 57

Table 5.1. Hydrogel compositions ...... 74

Table 5.2. Fluorophore propertiesa ...... 82

Table 5.3. Hydrogel Properties ...... 83

SI Table 5.1. Experiment summary table...... 101

Table 6.1. Values corresponding to Figure 6.1 ...... 118

xi LIST OF FIGURES

Figure 2.1. EO sampling of fluorescent beads ...... 9

Figure 2.2. EO sampling of Thioglo-1 ...... 10

Figure 2.3. EO cell loading of calcein AM ...... 12

Figure 2.4*. Fluid velocity as a function position ...... 13

Figure 3.1. The apparatus...... 21

Figure 3.2. Experimental sequence ...... 27

Figure 3.3. Intensity profiles ...... 28

Figure 3.4. Effective mobility in tissue as a function of the electrophoretic mobility ...... 29

Figure 3.5. A micrograph of an organotypic hippocampal slice culture ...... 31

Figure 4.1. The apparatus...... 50

Figure 4.2. Sample experimental run ...... 53

Figure S- 4.1. Circuit diagram of the differential amplifier in the four-electrode potentiostat. ... 64

Figure 5.1. Hydrogel cell ...... 78

Figure 5.2. OHSC experimental setup ...... 80

Figure 5.3. Two-pipette experiment ...... 84

Figure 5.4. Solute penetration distances ...... 87

Figure 5.5. Effect of current on concentration and Pe ...... 88

xii Figure 5.6. Steady-state PeE vs. PeT ...... 89

Figure 5.7. Injections into an OHSC ...... 90

SI Figure 5.1. Steady state versus non-steady state ...... 100

SI Figure 5.2. Comsol Simulation ...... 105

SI Figure 5.3. Effects of mismatched ζ-potentials ...... 106

SI Figure 5.4a. Digital MidgardTM experimental circuit ...... 108

SI Figure 5.4b. PAR experimental circuit ...... 108

Figure 6.1. Measured thicknesses over time ...... 117

xiii PREFACE

The successful completion of my PhD reflects the patience, encouragement, and intellectual

contributions of many individuals along the way. I am most grateful to my parents, Avital and

Shimon, who have supplied endless reassurance and strength. They showed me the value of

education and were my strongest champions. Deepest thanks go to my brother, Nir, who inspires me to become a happier, more fulfilled person every day. I want to express great appreciation for

Todd Bindrim, who supported, listened, and took an interest when I needed it most.

Many thanks go to the people who make up the Department of Chemistry at the

University of Pittsburgh. Thank you to each member of the Weber lab, both past and present.

You have been good friends and colleagues throughout the years. I would especially like to thank past members Moon Chul Jung and Imants Zudans for their mentorship, technical support, and friendship. To my incredible friend and colleague, Xiaomi Xu, I owe my sanity and motivation to persevere. Thanks to Ryan Stayshich for friendship and library study time. I would especially like to thank Amy Hamsher and Amir Faraji for being both good friends and superb colleagues on this project. Thanks go to Yanhong Yang for being a good friend and supporter. To Manyan

Wang, thank you for your efforts in working through the details.

I count myself very fortunate to have encountered some of the best hearted people in the

Department of Chemistry. Kind words of the staff within the department made the graduate school years possible, to which a special thank you goes to Michele Monaco. To Chuck

xiv Fleishaker, Dave Emala, and Bob Muha, your endless work has made this dissertation possible- thank you. A very sincere gratitude goes to Jeff Sicher and Tom Gasmire for their creativity and hard work, and mostly for their encouragement.

I am deeply grateful to my advisor, Prof. Stephen Weber, for his guidance and support, and for the many hours spent reading my manuscript and offering useful input. He allowed me to explore my scientific creativity and solidify my scholarly work. I am very grateful to the members of my committee, Prof. Adrian Michael and Prof. Shigeru Amemiya, who have supported me and have taught me to think carefully and purposefully. I owe my scientific bearing to Prof. Mats Sandberg, who introduced me to a new avenue of neuroscience research.

Thank you, Mats, for your extensive technical expertise on this project.

This dissertation is dedicated to my two grandmothers, who believed in the value of higher education and a woman’s ability to achieve her goals.

xv 1.0 INTRODUCTION

Studies of electric fields in the brain focus on the effects of endogenous electric fields,

environmental electric fields, and stimulation using electric fields for clinical and research

purposes. The brain creates endogenous electric fields in the process of communication between

neurons. Electric fields are naturally created in the synapse with ion channel openings.1

Furthermore, local field potentials occur where extracellular (EC) potential fluctuations contribute to synchronizing groups of firing neurons.2 Orientation of those neurons can further

amplify EC electric fields produced by synchronized firing.3

Some treatments of neurological diseases are based on the application of electric fields to

the brain. Deep brain stimulation is one such treatment that uses electric fields to treat the

physical effects of Parkinson’s disease, such as tremors.4 Transcranial magnetic stimulation5-8

and electric stimulation9-12 are clinical techniques that are used to treat depression and in stroke

rehabilitation. Furthermore, seizures are controlled with the application of electric fields in the

brain.13 Electric fields are also used to achieve electroporation in the brain in order to abolish

tumors14 or for genetic manipulation.15 In electrophysiology, electric fields are used to stimulate

single or small groups of neurons16 and even to direct neuronal growth17. Electric fields are used

in solute delivery methods, such as localized18-20 and transnasal iontophoresis21. Despite the

presence of endogenous electric fields and the use of exogenous electric fields, the degree to

which electric fields in the brain motivate EC solute transport is unknown.

1 1.1 ELECTROKINETIC TRANSPORT

An electric field applied across a porous medium, such as brain tissue, may produce

electrokinetic transport, depending on the nature of the solute and the EC environment. Cell

membranes and stationary EC matrix components include phospholipid head groups, proteins,

sulfated carbohydrates, and the like that ionize in the presence of an electrolyte solution such as

cerebrospinal fluid. An electric field in tissue with surface charges creates an electrokinetic bulk

fluid flow called electroosmotic flow (EOF). The charges on cell surfaces and fixed EC matrix

components affect the spatial distribution of ions in the EC fluid. A concentration gradient of

ions forms near the charged surface: counterions concentrate near the surface, and coions are in

low concentration.22 At a distance of 1 to 10 nm away from the charged surface, the ion

distribution in the bulk solution reaches electroneutrality. EOF arises by virtue of the charge

distribution in solution and the externally applied electric field.

The electric potential at the slip plane between the moving and stationary ions is termed

the ζ-potential.22 The magnitude of ζ-potential controls the magnitude of the EOF.

Electrosmosis, electrophoresis, and streaming current/ streaming potential are common

techniques used to experimentally determine the ζ-potential of a stationary surface in a porous medium or a conduit. EOF governs the observed velocity, vobs, of a neutral solute under the

influence of an electric field, E. Measuring vobs can be used to determine the ζ-potential, ζ, of the

surface. The velocity of the neutral solute is described by the Helmholtz-Smoluchowski

equation:

ζ = = = eo (1)

obs eo −휀 푣 푣 �휂 퐸 휇 퐸

2 where ε and η are the permittivity and viscosity of the medium, respectively. The electroosmotic

(EO) mobility, µeo, is directly related to the ζ-potential of the charged surface.

The ζ-potential of a particle can be determined in much the same way. The surface

charges on the particle define its ζ-potential. The migration of charged particles under an electric

field, independent of the ζ-potential of the matrix, is called electrophoresis. The ζ-potential of a

particle is related to its electrophoretic mobility, as described by Equation 2.

ζparticle = ep = (2)

ep 휀 푣 휇 퐸 �휂 퐸 The electrophoretic velocity, vep, is a property of the solute related to its mobility, µep. The net

migration of the solute is zero when µep directly opposes µeo. If µep is known, the ζ-potential of

the surface can be calculated using Equation 1.

Determination of ζ-potential using streaming currents includes a pressure gradient,

sometimes resulting from gravitational or centrifugal forces.23 Electrodes are placed on either side of the pressure gradient across the charged channel or porous matrix. The current at which the electrode potentials are equal is defined as the streaming current. The streaming current, istr,

is used to calculate ζ-potential by the following relationship:

ζ a2 P = L (3) 푠푡푟 휀 Δ 푖 �휂 � � � where a is the pore radius, ∆P/L is the pressure gradient between the electrodes.

3 1.2 PHYSICOCHEMICAL BRAIN PROPERTIES

ζ-potentials of isolated biological particles are often determined by measuring their electrophoretic velocities. ζ-potentials have been determined for isolated cells,24, 25 EC

constituents,26 and synaptosomes27. The ζ-potential of cartilage models composed of bone

fragments has also been investigated.28 In general, ζ-potentials of intact biological tissue have not been measured. However, EOF has been quantified in intact skin tissue to understand transdermal iontophoresis. Iontophoresis is an electrokinetic method used to drive solutes into tissue using an applied electric field. Research in transdermal iontophoresis has found that skin pores have an appreciable ζ-potential and the bulk of skin tissue has zero ζ-potential.29, 30 Ranck

estimated brain tissue ζ-potential to be approximately -15 mV by using previously determined ζ-

potentials of cell plasma membranes.31 This dissertation is mainly focused on the determination

of brain tissue ζ-potential and tortuosity and applying the knowledge to iontophoresis and

reverse iontophoresis (EO sampling).

EC solute transport is dependent on the geometry of the EC space and solute size and

shape. Outside of electrokinetic phenomena, solute transport over distances greater than the

synapse occurs by diffusion. EC diffusional transport has been thoroughly investigated.18

Tetramethylammonium ion (TMA+)32-58 , dextrans36, 59-62 and other polymers,63-65 albumin,60

radiotracers,66-69 neurotransmitters,33 and other molecules44, 61, 70-75 have been used as solutes for

diffusion experiments in the brain to investigate tortuosity. Tortuosity is a quantitative measure

of the twists and bends of the EC space. Various definitions of tortuosity exist. Volumetric

tortuosity is defined as a purely geometric length ratio: the effective distance a solute must travel

in the EC space between two points divided by the linear distance between those two points, i.e.,

4 as if there were no obstacles in the solute’s path.76 definition of volumetric tortuosity is valid

where the solute path is restricted to a single axis. The Multiple three dimensional pathways in

the brain, cannot be described by distance of the solute path. Therefore, neurobiologists define

tortuosity as apparent tortuosity. The square of the apparent tortuosity is equal to the ratio of a diffusion coefficient in free solution to the diffusion coefficient in the tissue matrix, i.e., apparent diffusion coefficient.18 The apparent diffusion coefficient, and thus apparent tortuosity, accounts for the three dimensional structure of the EC space, the solute properties, such as size and shape, and any frictional or viscous drag the solute experiences in the EC space.

Different areas of the brain have different values of tortuosity, and anisotropy can exist

within individual brain regions. The methods used to determine ζ-potential described in this

dissertation require solutes to travel through the EC space under the influence of an electric field,

and therefore subject solute movement to the tortuous environment of the brain.77, 78 We

designate this tortuosity electrokinetic tortuosity. Electrokinetic tortuosity describes the possible

dependence of solute flow on pore orientation with the electric field.79, 80 Published methods of

determining diffusional tortuosity in the brain measure diffusional tortuosity, absent of an

applied electric field.81, 82 Both ζ-potential and electrokinetic tortuosity of tissue can be defined

using the experimental methods described later.

1.3 THE MODEL SYSTEM: ORGANOTYPIC HIPPOCAMPAL SLICE CULTURE

In order to determine the ζ-potential of brain tissue, which had not been determined before the

research presented here, we need an appropriate experimental platform. Organotypic

hippocampal slice cultures (OHSCs) are used in electrophysiology and pharmacology

5 experiments.83 OHSCs serve as an in vitro model system for the brain for determining the ζ- potential and electrokinetic tortuosity. There are two different preparations of OHSCs: roller tube and interface. Both methods start with acute tissue slices that are extracted from specific brain regions and sliced at 100 to 500 µm thick. The roller tube preparation embeds the tissue slice in a plasma clot or collagen matrix on a microscope coverslip and the tissue is rotated slowly during incubation. This preparation yields a monolayer of cells that retains the cytoarchitecture of the original tissue slice. The interface preparation, originally developed by Stoppini, places slices on a porous hydrophilic PTFE membrane.84 The tissue is incubated over medium, such that nutrients from the medium permeate through the membrane into the tissue from below and oxygen is obtained from above. The initial thickness is critical for the interface method. Tissue slices larger than 500 µm are likely deprived of oxygen and nutrients. The interface method yields three dimensional cultures, about 4 cell layers thick, in which the cytoarchitecture and biochemistry are largely retained. This research employs the interface OHSC method.

OHSCs are ideally prepared with early postnatal rat pups, p0 to p10. At this age, the brain is easiest to dissect, brain regions are mostly established, and cells have a higher rate of survival following the surgical process. Some cell adaptability still exists, such that once in culture, the tissue slice recovers from the trauma of explantation with a degree of synaptic rearrangement.

Thus, complete tissue systems are established. Explantation from adult rats is possible, but requires oxygenation to avoid necrosis. Therefore, we chose to explant tissue from p6-p9 rats.

The hippocampus is the most widely studied organotypic culture using the interface technique. The hippocampus is part of the limbic system of the brain, and plays an important role in learning spatial navigation and long-term memory. The hippocampus is one of the main brain regions affected by Alzheimer’s disease, epileptic seizures, and ischemic conditions such as

6 stroke. Each OHSC contains the laminar structure of the hippocampus. Pyramidal cells make up

the cornu ammonus (CA), which is itself divided into 4 subgroups: CA1, CA2, CA3, and CA4.

The CA2 and CA4 are small relative to the CA1 and CA3 and are often not pointed out in

morphological studies. Granule cells make up the dentate gyrus. The dentate gyrus sits at the end of the CA4. Signals travel along the laminar structure from the dentate gyrus to the CA1. OHSCs also include the subiculum and entorhinal cortex. The CA1 dispatches signals to the subiculum.

Both the subiculum and entorhinal cortex subregions contain axons that project to other brain regions. The laminar structure of the hippocampus has been shown to generate endogenous EC

electric fields of tens of millivolts per millimeter by the synchronous activation of multiple

neurons.3

1.4 QUANTITATIVE ANALYSIS OF THE EXTRACELLULAR SPACE

Quantitative analysis of solutes in the EC space in living systems has proven difficult. EC

sampling can provide insight to EC solute concentrations, but the available methods bear

disadvantages. Microdialysis uses a probe to sample a broad spectrum of solutes, but the probe

damages the surrounding tissue.85, 86 Thus, results are a consequence of a perturbed system.

Furthermore, the time resolution of published microdialysis samples is on the minutes timescale.87 Other methods, such as chronoamperometry and fast-scan cyclic voltammetry

(FSCV) use microelectrodes that are small in diameter, avoiding cell death as a result of probe

placement. FSCV has millisecond time resolution, but can only detect changes in levels of

electrochemically active species.87-89 Micropush-pull sampling uses induced pressure to sample

7 from the EC space. Disadvantages of this sampling method include sample dilution, mixing, and inferior time resolution to FSCV.90-92

Ranck’s estimation of a brain tissue ζ-potential of -15 mV31 leads to the possibility of motivating electrokinetic flow in the EC space of the brain via an applied electric field. Provided a significant ζ-potential, such as -15 mV, electrokinetic flow can be directed within the brain.

This implies that EOF could be therefore be directed from the brain and into a conduit such as a sampling capillary for separation and analysis of the EC contents.

8 2.0 INITIAL MOTIVATION

We have proposed an EC sampling method using the principles of electrokinetic forces. The

method would include the advantages of sampling a broad spectrum of analytes and short time

resolution, without damaging cells in the sampling region. The basic principles of this sampling

idea, termed EO sampling, were tested by sampling fluorescent beads into a fused-silica

capillary, as shown in Figure 2.1.

Figure 2.1. EO sampling of fluorescent beads

A: A 50 µm inner diameter (30 cm long) fused-silica capillary, filled with electrolyte solution, is placed in a suspension of 2 µm diameter fluorescent beads over an inverted fluorescence microscope objective lens. The

distal end of the capillary is in a vial containing electrolyte solution and a platinum electrode. A circuit is

created such that the electric field emanates in the bath and into the capillary. B: Fluorescent bead position tracking shows the beads moving into the lumen of the capillary with the application of voltage.

Figure 2.1 displays the tracked positions of the fluorescent beads over a period of time in which

an electric field is applied. The beads are transported into the capillary lumen by virtue of EOF

9 produced in the capillary. Therefore, Figure 2.1 shows that we are able use EO sampling to

sample from an electrolyte bath, into a fused-silica sampling capillary.

By knowing the overall ζ-potential of brain tissue and applying an electric field through

the brain, we can direct EOF within the EC space. EOF can be directed from within the brain

into a conduit to sample electroosmotically. Furthermore, we can direct fluid flow from the

conduit into the brain as a possible application for drug delivery. EOF also could be useful for encouraging volume transmission within the EC space. Figure 2.2 demonstrates EOF directed from below an OHSC membrane, through the OHSC, and into a sampling capillary.93

t = 0 s t = 6.6 s t = 13.2 s

Figure 2.2. EO sampling of Thioglo-1

The injection capillary, filled with Thioglo-1, is placed over a gold minigrid and under the OHSC membrane.

The edge of the injection capillary is indicated by a green line. A sampling capillary is placed orthogonally to the OHSC. The lumen of the sampling capillary is emphasized by the dotted circle in the top row, where it was otherwise not obvious. The top row displays the injection of Thioglo-1 in the case where no voltage is applied between the gold minigrid and the sampling capillary. The bottom row displays the progression of a similar injection of Thioglo-1, while 3 kV are applied. An inverted fluorescence microscope (4X objective lens) imaged the experiments.

10 The control experiment in the top panel shows the pressure ejection of Thioglo-1 under the

OHSC membrane. Thioglo-1 fluoresces in the presence of cell thiols, mainly glutathione. Cells near the ejection capillary (indicated by the green line) fluoresce over time as Thioglo-1 diffuses into the OHSC. By contrast, application of an electric field through the OHSC and into the sampling capillary, as shown in the bottom panel of Figure 2.2, directs Thioglo-1 into the OHSC under the sampling capillary lumen by virtue of EOF. Thus, this set of experiments shows that the principles EOF can be used for EC sampling.

Interestingly enough, the same basic experiment can be used to measure enzyme kinetics.

Such is the example with calcein acetoxymethyl (AM) ester shown in Figure 2.3. The top series shows the control experiment in which calcein AM diffuses from below the OHSC membrane and into the OHSC without an applied electric field. The OHSC does not fluoresce after 15 minutes, indicating calcein AM has not diffused into the cells. The second series shows the same setup under the influence of an electric field. After 15 minutes, the cells under the sampling capillary fluoresce. Uncharged calcein AM enters living cells, and esterases within the cell hydrolyze calcein AM generating a fluorescent green calcein complex.94 Calcein AM hydrolysis generally takes 15 minutes to 1 hour.95, 96 Therefore, Figure 2.3 indicates EOF of calcein AM occurred almost instantly with the application of electric field. Cell death analysis of OHSCs confirms no additional cell death occurs as a result of EO sampling under controlled conditions.97

Figure 2.3. EO cell loading of calcein AM

The images are taken at 5 minute intervals using an inverted fluorescence microscope. The fluorescent injection capillary filled with calcein AM lies under the OHSC membrane. We replaced the gold minigrid used in experiments described earlier (refer to Figure 2.2) with a platinum electrode. The top series shows the control experiment where calcein AM has been injected under the membrane without application of voltage.

The bottom series shows the application of 1.8 kV following calcein AM injection. After 15 minutes, cells in the bottom panel fluoresce. The bottom image shows a bright field image of the OHSC with the injection capillary under the OHSC membrane and the sampling capillary orthogonal to the OHSC.

Using capillary electrophoresis, the ζ-potential of a fused-silica capillary filled with physiological buffer, such as those shown in the previous three figures, was found to be approximately -50 mV - more than double that of an OHSC.77, 78 Despite the ζ-potential mismatch at the capillary-OHSC interface, simulations of EO sampling show a relatively low pressure generated at the interface of the capillary and the OHSC under fairly aggressive

conditions (-2000 V, with a 30 cm-long 100 µm inner diameter capillary) however there is

nonetheless a pressure-induced flow generated. Three dimensional Comsol simulations using the

geometry described above show that the fluid velocity in the sampling capillary is lower due to

pressure created by a ζ-potential mismatch (refer to Figure 2.4*) in comparison to the case where

there is no mismatch. On the other hand, an increase in fluid velocity occurs at the same time

within the simulated OHSC.

0.30 0.25 0.20 0.15 Sum 0.10 Pressure 0.05 Field Velocity mm/s 0.00 -0.05 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Position, mm

Capillary-OHSC Interface

Figure 2.4*. Fluid velocity as a function position

*Courtesy of Dr. Imants Zudans. Fluid velocity as a function position is separated into two components: the

pressure-induced velocity (black line), and the EO velocity (shown as a red dashed line). The sum of the two

velocities (shown in blue) is the observed fluid velocity. The capillary ζ-potential is set to -50 mV and the

OHSC ζ-potential to -15 mV. The OHSC is 300 µm thick, hence the interface is at position 0.3 mm on the plot. A schematic of the setup is shown to the right of the plot.

In this chapter, we were able to prove that EOF can be produced in an OHSC. Moreover,

EOF can be directed from the OHSC into a sampling capillary. Pressure-induced flow during

EOF sampling likely exists due to a ζ-potential mismatch between the OHSC and the sampling

13 capillary. In order to quantitatively understand EO sampling from the OHSC EC space, we need to know the ζ-potential of the OHSC.

14 3.0 DETERMINATION OF ZETA POTENTIAL IN RAT ORGANOTYPIC

HIPPOCAMPAL CULTURES

This work is published in Biophysical Journal 2008, 94 (11), 4561-4569.

ABSTRACT

ζ-potentials of entities such as cells and synaptosomes have been determined, but ζ of brain tissue has never been measured. Electroosmotic flow, and the resulting transport of neuroactive substances, would result from naturally occurring and experimentally or clinically induced electric fields if ζ is significant. We have developed a simple method for determining ζ in tissue.

An electric field applied across a rat organotypic hippocampal slice culture (OHSC) drives fluorescent molecules through the tissue by both electroosmotic flow and electrophoresis.

Fluorescence microscopy is used to determine each molecule's velocity. Independently, capillary electrophoresis is used to measure the molecule’s electrophoretic mobilities. The experiment yields ζ-potential and average tissue tortuosity. The ζ-potential of OHSCs is –22 ± 2 mV and the average tortuosity is 1.83 ± 0.06. In a refined experiment, ζ-potential is measured in various subregions. The ζ-potentials of the CA1 stratum pyramidale, CA3 stratum pyramidal and dentate gyrus are -25.1 ± 1.6 mV, -20.3 ± 1.7 mV and -25.4 ± 1.0 mV respectively. Simple dimensional

15 arguments show that electroosmotic flow is potentially as important as diffusion in molecular

transport.

3.1 INTRODUCTION

An understanding of brain function, especially extracellular transport, requires a quantitative

understanding of the physicochemical properties of the extracellular space. Diffusion in the

extracellular space of the brain and slice preparations is fairly well understood. Nicholson98, 99

and Sykova100 measured the transport of tetramethylammonium ion (TMA+) and polymers to

reveal the details of extracellular diffusion. Other methods such as cortical surface

photobleaching,101 magnetic resonance,102-105 integrative optical imaging,106 and dual-probe

microdialysis107, 108 have been used to evaluate the diffusive transport of markers and

biomolecules as well as water in living brain tissue. A detailed understanding of variations in

diffusion caused by changes in extracellular volume fraction has led to applications of these

techniques in assessing, e.g, edema.101 The dimensions of the intercellular space have been

determined by measuring particle diffusion rates.109 Tortuosity has a large effect on transport

rates in heterogeneous media. Tortuosities vary depending on the local geometry of the tissue.110,

111 In addition, certain brain regions are anisotropic112 leading to a valuable method in magnetic

resonance imaging called “diffusion tensor imaging”.102, 105 We conclude that there is a rather

firm understanding of diffusive transport in the brain, which is critical for understanding

extrasynaptic (paracrine or volume) transmission,113, 114 neuronal development115 as well as for

making inferences from measurements.89, 116

16 On the other hand, there is less known about electrically mediated transport in the brain.

Electric fields and currents in heterogeneous media create a fluid flow, called electroosmotic

(EO) flow, by virtue of the field’s motivating force on the mobile counterions to fixed charges on

surfaces or immobile macromolecules in the extracellular space. The fixed charges create a ζ-

potential, the potential at the shear plane between a charged surface and a moving electrolyte

solution22. The magnitude of the ζ-potential controls the magnitude of the EO flow. The ζ-

potential in brain tissue is a consequence of cell-surface functional groups, such as phospholipids, proteins, and carbohydrates as well as fixed components of the extracellular matrix.

The focus of this work is to create and apply a method for the determination of the ζ- potential in nervous tissue. In the paragraphs below, we justify the need to understand this currently undetermined parameter of brain tissue, and demonstrate how a method similar to electrophoresis can be used.

Certainly, electric fields naturally occur in brain. Recently, Savtchenko et al. showed how the electrical field created in the synapse during ion-channel opening can influence molecular transport in the synapse, normally viewed as occurring by diffusion.117 Several clinical

techniques such as transcranial direct current stimulation118-121 and magnetic stimulation,122-125

also create electric fields in brain.126 These treatments are effective in depression,120, 123 stroke,119

and other diseases. Electric fields are also used in seizure control.127 In research laboratories,

electric fields are applied to brain tissue to direct neuronal growth,17 to perform iontophoresis in

the brain,128 carry out electroporation in brain15 and embryos,129, 130 and for transnasal delivery of

a charged peptide (by an electric field created between the nasal cavity and the back of the

head21). As far as we are aware, EO flow is largely overlooked in these endeavors.

17 EO flow and/or ζ-potential have been measured in some biologically relevant matrices by

a variety of techniques. EO flow is knowingly created in transdermal iontophoresis.30, 131-136 EO

dewatering of tumors also depends on the existence of a ζ-potential within tissues.17, 137 Outside

the fields of transdermal iontophoresis and transdermal sampling (reverse iontophoresis136, 138),

measurements of ζ-potential, or EO mobility, are rare in biological systems. Values of ζ-

potential have been reported for isolated cells using electrophoresis,139 microelectrophoresis,140

and microcapillary electrophoresis on a chip.141 Electrophoretic methods have also been applied

to bone particles142 as well as isolated nuclei,143 vesicles,144 and synaptosomes145 from brain

tissue. Measurements of transport through tissue, such as skin that can act as a film separating

two liquid phases, are based on quantitative determination of the flux of a solute from a “source”

phase to a “receiving”’ phase. Similarly, EO mobility in plant tissues146 can be made this way.

Alternative measurements using an oscillating electric potential,147 or scanning electrochemical microscopy on individual skin pores are also used.30 None of these methods is suited to whole

brain, acute brain slices, or slice cultures.

At a more fundamental level, knowing the ζ-potential in intact tissue is vital to understanding molecular movement between cells. Electroosmosis has been proposed as a contributor to learning and memory31 Even local electric fields, such as those that materialize

during epileptic seizures, may generate fluid motion. In tissues other than brain, i.e., lung148 and

epithelial tissue,149 naturally occurring potential gradients are thought to drive fluid flow. A

mathematical theory, including double layer overlap and coupling of mechanical and electrical

forces, has been applied to glycogen-containing cartilage models.28

We conclude that EO phenomena have been studied in tissues where it clearly must

contribute to the observed phenomena, e.g., skin (iontophoretic drug delivery and sampling) and

18 bone (stimulated bone growth). However, there is virtually no understanding qualitatively, and

certainly no understanding quantitatively of the ζ-potential in nervous tissue.

We have developed a method to measure EO mobility, thus ζ-potential as well as average tortuosity in organotypic slice cultures of the rat.84 The velocities of fluorescent molecules are

measured in the cultured tissue under an applied field. Data analysis is guided by simple theory,

shown below.

3.1.1 Theory

Consider a set of fluorescent molecules with various electrophoretic mobilities. A spot consisting of a single fluorophore is placed in an OHSC. In an electric field, the fluorophore will

move at an observed velocity, vobs, as given in Eq. 1, which can be inferred from Boyack and

150, 151 Giddings(62) and Rathore et al.(63) (λε=tortuosity; veo=electroosmotic velocity; vep=

electrophoretic velocity):

 1  v obs =  2 (v eo + vep ) (1)  λe 

We assume that veo and vep experience the same tortuosity. Note that the definition of tortuosity

is consistent with Boyack and Giddings150 and Rathore et al.151 We denote that the tortuosity

influences an electrokinetic process with the subscript “e”. This tortuosity may not be the same as the ratio of free and effective diffusion coefficients (see Discussion for more detail).

Equations 2 and 322, 151 define the velocities as the products of mobilities, µ, and a field, E.

vep = µep E (2)

εζ v = − E = µ E (3) eo η eo

19 Equation 3 shows that µeo, is a function of ζ, the ζ-potential, the medium’s viscosity (η) and permittivity (ε). Values of µep can be determined for each fluorophore independently in

HBSS by CE; thus they are known quantities. The observed mobility, µobs, can be defined as Eq.

4 for an individual fluorescent species. If the tortuosity is not dependent on the fluorophore, a plot of µ versus µ will yield a straight line with a slope of 1 where the brackets indicate obs ep λ2

“average”. The intercept provides µeo, and thus ζ.

v   εζ    obs  1   1  µobs = =  2  µep −  =  2 (µep + µeo ) (4) E  λe  η   λe 

Another approach involves the use of two molecules simultaneously (molecule 1 and molecule 2). Each molecule is independently described by Eq. 5. Thus the term λ2/E can be eliminated, as shown in Eqs. 6 and 7 (subscripts 1 and 2 refer to individual molecules).

v λ 2 µ = obs e − µ (5) eo E ep

 v  µ =  obs1  µ + µ − µ eo  ( eo ep2 ) ep1 (6)  vobs2 

 v   obs1   (µep2 − µep1 )  vobs2  εζ µeo = = − (7)  vobs1  η 1 −    vobs2 

20 3.2 MATERIALS AND METHODS

3.2.1 Chemicals and solutions

The following materials were purchased from Sigma (St. Louis, MO) and were used as received.

The culture medium contained basal medium Eagle (50%), Earl’s basal salt solution (20%), horse serum (23%), penicillin/streptomycin (25 units/ml), L-glutamine (1 mM), and D-(+)-

glucose (7.5 g/L). The culture medium was stored in a refrigerator and warmed to 37°C before

use. Gey’s Balanced Salt solution (GBSS) was supplemented with 27.5 mM D-(+)-glucose and

2.7 mM MgSO4. The following solutions were prepared with 18 MΩ purified water from a

Millipore Synthesis A10 system (Millipore, Billerica, MA): NaOH solutions, HEPES-buffered

salt solution (HBSS) containing (mM): 143.4 NaCl, 5 HEPES, 5.4 KCl, 1.2 MgSO4, 1.2

NaH2PO4, 2.0 CaCl2, and 10 D-(+)-glucose. HBSS was filtered, stored in the refrigerator and

warmed to 37 °C before use. Approximately 3% agarose type-VII in HBSS filled the apparatus shown in Fig. 3.1.

Figure 3.1. The apparatus

The OHSC and its insert membrane are placed against the cast agarose gel. The apparatus is placed on an inverted fluorescence microscope for data acquisition. Reference electrodes measure the field strength near the fluorophore spots.

21 Each of the fluorescent molecules was diluted with HBSS, filtered, and frozen until use.

The concentration of each probe was empirically determined to accommodate its intensity under experimental conditions. See Table 3.3 in Supporting Material, for the concentrations.

Fluorescent dextran conjugates 1-6 were purchased from Invitrogen/Molecular Probes (Eugene,

OR). 7 and 8 were obtained from Sigma (Table 3.1).

Table 3.1. Values of µep in HBSS

µ /(10-9 SEM/(10-9 Fluorescent Compound ep m2/Vs) m2/Vs) ‡ AlexaFluor 568 dextran conjugate 10 kDa (1) -1.92 ± 0.15 BODIPY FL dextran conjugate 10 kDa‡ (2) -0.89 ± 0.22 Fluorescein dextran conjugate 70 kDa‡ (3) -9.75 ± 0.82 Oregon Green dextran conjugate 10 kDa‡ (4) -18.0 ± 0.1 Rhodamine Green dextran conjugate 3 kDa‡ (5) -1.72 ± 0.19 Texas Red dextran conjugate 70 kDa‡ (6) -0.55 ± 0.19 Fluorescein sodium salt (7) -26.3 ± 0.3 tris(2,2’-bipyridine)ruthenium (8) 26.0 ± 0.0 ‡Dextran conjugate

3.2.2 Instrumentation

3.2.2.1 Capillary electrophoresis:

Electrophoretic mobilities in HBSS were determined using an ISCO Capillary Electropherograph

(Lincoln, NE) with vacuum injection (10 s, 0.5 psi) and ultraviolet-visible absorbance detection.

A section of the polyimide coating was removed with a low-temperature flame to create the

window. New capillaries were flushed with base overnight. Before each series of experiments,

capillaries were flushed with 200 µL of distilled, deionized water, followed by the same volume

of the running buffer. The high voltage was then applied to the capillary, and the system

equilibrated for 5-20 min. All injections included rhodamine B as the EO flow marker, and were

22 performed at ambient temperature (23°C ± 2°C). The EO flow marker is a neutral molecule with

a mobility equal to zero. Its velocity in a capillary electrophoresis (CE) experiment is the

electroosmic velocity.

CE measurements of mobilities were performed using 360 μm outside diameter (OD)

capillaries with inner diameters (IDs) of 40 or 50 μm. The total capillary length ranged from 50.3

to 118.4 cm, with 25.5-48.1 cm from the injection end to the detection window using -9.0 to -

22.1 kV. Detection was at 495, 505, or 595 nm. See Table 3.4 for the parameters used for each

particular molecule.

3.2.2.2 Apparatus for tissue experiments

The apparatus in Fig. 3.1 was manufactured from Lucite locally. A 40-V direct current power

supply (Heath, Benton Harbor, MI) applies a low homogeneous field (~22 V/cm at the tissue)

across the organotypic hippocampal slice culture (OHSC). Two Goodfellow (Cambridge,

England) tubular platinum electrodes (1 mm OD, 0.1 mm wall thickness, and ~3.8 cm long) were placed into separate cylindrical reservoirs (19 mm in diameter and 12 mm deep) separated by a narrow rectangular channel 44 mm in length. These reservoirs are not shown in Fig. 3.1. The

OHSC remained on the insert membrane during the measurements. Electrical contact between

the OHSC and the electrodes was made by agarose-containing HBSS.

3.2.2.3 Ag/AgCl electrodes

A pair of in-house fabricated Ag/AgCl microelectrodes measured the potential gradient in the

tissue. The electrodes were made from 0.1-mm-diameter silver wire from JMC Ltd.

(Hertfordshire, England). Two sets were made: one spaced 0.80–0.90 mm apart and another

23 spaced 0.35–0.40 mm apart. Distances were determined using a scale bar and micrographic images of each individual experiment.

3.2.2.4 Injection capillary

A 250 µm (ID) fused-silica capillary was pulled to a tip (~12 µm OD) using a P-2000 capillary puller from Sutter (Novato, CA). A Sutter MP-285 micromanipulator lowered the capillary into the OHSC. The distal end of the capillary was attached to a Harvard PHD 4400 Hpsi syringe pump (Holliston, MA) with a 50 µL Hamilton Gastight syringe (Reno, NV). The pump delivered fluid at a flow rate of 120 nL/min for ~1 s.

3.2.2.5 Imaging

An inverted fluorescent microscope (Model IX71) with a U Plan Apo 4× objective lens (both from Olympus, Melville, NY) imaged the tissue experiment using a charge-coupled device camera (ORCA-285 Hamamatsu, Hamamatsu City, Japan). Depending on the fluorescent properties, an appropriate cube was chosen. The Olympus fluorescence cubes are a wide interference filter blue cube (exciter 460–495 nm, dichromatic mirror 505 nm, emitter 510-550 nm interference filter) and wide interference filter green cube (exciter 530–550 nm, dichromatic mirror 570 nm, emitter 575 nm interference filter). A Semrock (Rochester, NY) DA/Fl/TA-3X-

A Triple-band “Pinkel” filter set (exciter 1 387 nm, exciter 2 494 nm, exciter 3 575 nm, dichromatic mirror: 394–414 nm, 484–504 nm, 566–586 nm, emitter: 457, 530, 628 nm) was used for multiple fluorophore experiments.

Image processing and distance measurements were done with SimplePCI 6.0 software

(Compix, Cranberry, PA). Images of the paired Ag/AgCl electrodes using the microscope

24 camera were captured daily to measure the distance between the electrodes. The same was done

for measuring IDs of new injection capillaries.

3.2.3 Procedures

3.2.3.1 Hippocampal organotypic slice cultures

The Stoppini culturing method was used with small variations.84 The procedures described here

have been approved by the University of Pittsburgh IACUC. The hippocampi were dissected

bilaterally from decapitated, 7-day postnatal Sprague-Dawley albino rats. The hippocampi were

cut transversely into 500 μm thick slices with a McIlwain tissue chopper from The Mickle

Laboratory Engineering (Surrey, England). The cultures were placed on Millicell membrane

inserts from Millipore and incubated over culture medium at 95% air/5% CO2. Cultures may be

incubated for up to 2 weeks. All data reported here are from cultures incubated for 3–7 days. The

mean slice thickness in this time period was 148 µm consistent with other reports.84, 152, 153 On

the day of the experiment, 37°C GBSS replaced culture medium, followed by 30 min of tissue

incubation. This procedure was repeated once more. Lastly, 37°C HBSS replaced GBSS before

the experiment.

Using an X-Acto knife, the insert membrane was cut to liberate the tissue and the insert membrane under it from excess insert membrane. Edges of the insert were left to allow moving the OHSC into the apparatus without damage as in Fig. 3.1. A portion of the agarose gel was cut away to make room for the OHSC. The OHSC was placed into the apparatus in such a way that the insert edges are under the agarose, but the slice is adjacent to the gel, as shown schematically

in Fig. 3.1.

25 3.2.3.2 ζ-potential determination

The tip of the injection capillary, preloaded with fluorescent probe-containing solution at room temperature, was inserted approximately midway (~75 - 150 µm) into the OHSC with the aid of a micromanipulator. The pump delivered ~2 nL of the fluorescent material. On occasion, placing the capillary into tissue alone delivered enough fluorophore by capillary effects. The capillary was carefully pulled out of the slice, and the microelectrodes were placed in the tissue. Before the voltage supply is turned on, the exposure time is set using autoexposure and the SimplePCI

6.0 software begins recording the fluorescence. The voltage supply was switched on, applying

~22 V/cm through the slice. Measurements of electric field and velocity are taken once the electric field stabilizes (a few seconds). Images were acquired once per second. Movement of a fluorophore toward the cathode is defined as positive.

3.2.3.3 Two-fluorophore experiments

Two sets of paired molecules were used. One set was 1 and 7, and the other set was 3 and 8. To

measure the velocity of each fluorophore simultaneously, the “Pinkel” filter set was used. At

first, each fluorophore is injected into an OHSC separately. The separate exposure times are set

using the autoexposure command for a two color image in the SimplePCI 6.0 software. A

mixture of the paired fluorophores is then injected into a different OHSC for measurement.

3.3 RESULTS

Table 3.1 shows the CE-determined mobilities for each fluorophore. Fig. 3.2 shows images from

an experiment in an OHSC. The OHSC received three injections, labeled A, B and C. A and C are

26 injections into the insert membrane, and B is in the OHSC. We find that fluorophore injected into the insert membrane is immobile regardless of the selected fluorophore or applied field. Each point in Fig. 3.2 (1-8) functions as an intensity detector. We measure the intensity at each point

as a function of time (Fig. 3.3).

Figure 3.2. Experimental sequence

Alexa Fluor 568 10 kDa dextran conjugate was injected into the insert membrane (injections A and C) and

into the tissue culture (injection B). After applying the potential, injection B migrates toward the cathode, as

shown in the sequence of images. Injections A and C remain in place despite the electric field. Intensities are measured at points 1-8 as a function of time. The smaller injections A and C are ~135 µm in diameter.

Injection B is ~500 µm in diameter.

vobs corresponds to the slope of a plot of the relative positions of the detection points versus the peak

times (Fig. 3.3, inset). This plot routinely yields straight lines, indicating that fluorophore velocity and electric

field are constant during the experiment. Dividing the velocity by the applied electric field, measured using an

Ag/AgCl reference electrode pair placed in the tissue, yields the observed mobility (µobs) as shown in Eq. 4.

Figure 3.3. Intensity profiles

The intensity profiles recorded at each of the eight detection points on the cathodic side of the original

injection. Inset: The relative position of the detection points (1-8) in Fig. 3.2 are plotted as a function of the times of the peak maxima (inset). The linearity implies that the fluorophore's velocity is constant throughout the experiment.

Figure 3.4. Effective mobility in tissue as a function of the electrophoretic mobility

The line is the best fit to all of the data (linear regression). The labels at each point identify the identity of the fluorophore (see Table 3.1) and its molecular weight.

Fig. 3.4 is a plot of µobs versus µep. The errors are the mean ± SE and are also given in

Table 3.2. Analysis of the results (linear regression) according to Eq. 4 leads to Eq. 8.

µ − obs = 0.30(± 0.02)µ + 5.2(± 0.3)×10 9 (8) m 2 ep ( Vs)

The slope of 0.30 ± 0.02 is equal to the inverse square of the average tortuosity. The average

(electrokinetic) tortuosity is therefore 1.83 ± 0.06. The ζ-potential in the OHSC is -22 ± 2 mV.

29 Table 3.2. A summary of the average observed mobilities in tissue and their SEMs

µ /(10-9 SEM/(10-9 Fluorophore obs n m2/Vs) m2/Vs)

1 6.46 ± 0.67 58 2 4.09 ± 0.39 19 3 1.85 ± 0.24 22 4 0a - 14 5 4.31 ± 0.35 13 6 3.66 ± 0.71 12 7 -3.15 ± 0.39 55 8 10.3 ± 1.3 10 aNo measureable movement n is the number of experimental runs for that specific fluorophore.

The zeta potential described above is an average from all areas of the OHSC in the direction of the line, as indicated in Fig. 3.5 (approximately along the medial-lateral axis). As noted in the introduction, there are regional differences in the hippocampus. Are there different

ζ-potentials in different regions? A micrograph of an OHSC (Fig. 3.5) shows areas interrogated with the two-fluorophore experiments; 1 was paired with 7 in 41 experimental runs. The 3 and 8 pair was run twice.

The CA1 stratum pyramidale, CA3 stratum pyramidale, and dentate gyrus have ζ- potentials (respectively) of (mV) -25.1 ± 1.7 (n = 12), -20.3 ± 1.6 (n = 14), and -25.4 ± 1.0 (n =

34). Analysis of variance reveals that there is a significant difference among the regions (p <

0.05), which is accounted for by the difference between the value in CA3 and the other two.

Figure 3.5. A micrograph of an organotypic hippocampal slice culture

The circumscribed areas represent the portions of the CA1, CA3, and dentate gyrus (DG) investigated. The horizontal line represents the direction in which the tortuosity is measured in the single-fluorophore experiment.

3.4 DISCUSSION

3.4.1 Analysis of the method

CE measures the electrophoretic mobility of the fluorophores used in the tissue experiment. As a result of the high electrolyte concentration, we had to modify typical CE procedures. To minimize the effect of thermal gradients, a narrow (40 µm ID) and long (~100 cm) capillary replaces the more common 75 µm ID, 50-cm-long capillary. With the longer, narrower capillary, peak shape is good.

31 Labeled dextrans (Supporting Materials) are the most successful category of fluorophores for our experiments. Dextran fluorophores are easily visible with a fluorescence microscope; they have minimal “nonspecific” binding;98, 154, 155 and their CE results are unambiguous.

Apparently, the fluorescent labels on each of the dextrans account for their different mobilities.7 and 8, both nondextran conjugate fluorophores, also have similar characteristics, thus are acceptable for our study. Several other fluorophores, on the other hand, do not meet the necessary criteria (Supporting Materials).

Although an in vivo model preserves the true environment of tissue, several advantages make the OHSC the optimal model for our purposes. We can easily track temporal changes in tissue slice cultures using an inverted microscope. Moreover, tissue slices are viable models because the cytoarchitecture and electrophysiology are preserved after explantation.98, 151-153, 156-

158 Additionally, cultured slices are easier to manipulate than acute slices.

A tissue culture is easily placed in the apparatus as shown in Fig. 3.1. We use the apparatus to apply a potential gradient along the tissue by providing electrical contact through agarose gel. The design of the apparatus is such that the tissue sits in a narrow channel between two reservoirs, thus increasing the resistance in the channel. The increased resistance amplifies the electric field. The change in depth from the agarose gel to the OHSC further concentrates the field in the tissue. Thus, the field is largest within the OHSC. Applying a large field through a tissue can alter its permeability, or electroporate the cells.159 Typical fields in electroporation, ~1 kV/cm, are ~45 times larger than the field strengths applied in the apparatus. To isolate the experiment from any electrolysis products created at the platinum electrodes, the electrodes are placed in deep reservoirs far from the tissue. Early designs of the apparatus had much smaller reservoirs. pH-responsive fluorophore experiments showed that electrolysis products moved into

32 the channel within 6 min and altered the properties of the gel. The same pH-responsive

fluorophore experiments repeated in the current apparatus design show that the electrolysis

products remain in the reservoirs for well over 30 min. This provides plenty of time to run the

tissue experiment.

To preserve the integrity of the tissue while injecting fluorescent material, several

provisions are taken. Primarily, the injection capillary is pulled to a 12 µm OD tip to minimize

tissue damage. A micromanipulator allows control over the depth of the injection. The injection

capillary delivers ~2 nL of fluorophore into the tissue to maintain the microenvironment. There

is no obvious damage to the tissue as a result of injection.

The membrane insert has completely different properties than tissue. The membrane itself

is made of poly(tetrafluoroethene) with a proprietary hydrophilic surface modification. When

fluorescent material is injected into the membrane it appears to stay in place (Fig. 3.2, injections

A and C). Therefore, fluorescent compounds moving in the visual field are in fact injected into the tissue and not into the insert membrane.

3.4.2 Analysis of the results

The measured ζ-potential is -22 ± 2 mV. The value is in the range of isolated biological

materials, such as mouse N-18 cells measured by a microelectrophoretic apparatus with a ζ-

potential of -15 mV,140, 160 chromaffin granules, -14.4 mV,144 and rat brain synaptosomes using

particle electrophoresis, -11.8 mV.145 Theoretical calculations determined a cartilage model containing glycogen to have a ζ-potential of -27 mV.28 Ranck estimated the ζ-potential in brain

to be -15 mV based on the constituents of the plasma membranes of neurons and glial cells.31

33 The ζ-potential determined using two fluorophores simultaneously is independent of the

average tortuosity factor and field (Eq. 7). The mean ζ-potential from the simultaneous

fluorophore experiments is essentially the same as that in the single fluorophore experiments. We

infer from this either of two things: the values of the field and tortuosity from the single-

fluorophore experiments are correct, or there is a fortuitous cancellation of deterministic errors.

We view the latter conclusion as a remote possibility. Thus, the two-fluorophore-experiment further validates the experimental method.

The two-fluorophore method is capable of distinguishing small differences in ζ-potential between different structures in the brain: CA1 (-25.1 ± 1.7 mV), CA3 (-20.3 ± 1.6 mV) and dentate gyrus (-25.4 ± 1.0 mV). The CA1 and dentate gyrus regions have statistically the same ζ- potential at a 95% confidence level. However, a single-factor analysis of variance analysis reveals that there is a statistical difference among the three areas (p < 0.05). It is interesting to note that these regions have different vulnerabilities to excitotoxicity correlated with the differences in ζ-potential. The CA1 and dentate gyrus regions are more susceptible to N-methyl-

D-aspartic acid excitotoxicity than the CA3 region.161 Perhaps the ζ-potential differences between the regions take part in mechanistic differences.

The experiments yield an average (electrokinetic) tortuosity. Tortuosity in the neurochemical context as determined through diffusion measurements is defined as

1/ 2 λD = (D / Dobs ) where D is the molecular diffusion coefficient in an unobstructed medium, Dobs

is the observed diffusion coefficient, and the subscript D denotes diffusion. It is noteworthy that

tortuosity influences fluid flow22 and electrophoresis150 through porous media. Boyack considers

two contributions to tortuosity in electrophoresis: one is purely geometrical- the ratio of the actual distance that a molecule travels in its path around obstacles to get from point A to point B

34 to the straight-line distance between A and B- whereas the other is related to the amplification of

the electric field in narrow conduits. Similarly, tortuosity as measured by diffusion is related to a

geometrical component and other factors, e.g., local viscosity.106 Thus measurements made by

diffusion and by electrokinetic experiments may not yield the same result.

In the development of the data treatment (Eq. 4), we assumed that the tortuosity (as

determined by electrokinetic experiments) is independent of brain region and fluorophore size

and charge. In fact, there is a dependence of tortuosity (as measured by diffusion) on molecular

mass and type of polymer (in acute cortical slices).98, 162, 163 Using a small probe molecule,

TMA+, differences in tortuosity (as measured by diffusion) in different regions of the

hippocampus have been found.110, 112, 164 In cortical slices, Nicholson found that the range of

tortuosities for fluorescein conjugated dextrans in the same molecular mass range that we used

(3-70 kDa) was from 1.77 to 2.25.163 Thus, to the degree that the tortuosities measured

electrokinetically and diffusionally are the same, our measured tortuosity has to be considered an

average over the molecular masses and the brain regions studied.

If there is a correlation between the observed mobility and fluorescent species molecular

mass, molecular mass effects on tortuosity may lead to a deterministic error in ζ. If there is no

correlation, differences in tortuosity will add to the error in ζ, but will not affect its magnitude.

From Fig. 3.4, it can be seen that there is no significant pattern of deviation from the line due to molecular mass. The correlation coefficient of the molecular mass and electrophoretic mobility

(data in Table 3.1) is -0.07 demonstrating the lack of correlation. Thus, the assumption of an average tortuosity adds to the error in ζ, but does not influence the measured magnitude of ζ.

Furthermore, recall that the two-fluorophore experiment yields a result that is independent of

tortuosity (if it is again assumed that one tortuosity applies to both dyes). The average ζ-potential

35 from these results, as mentioned above, is indistinguishable from the value determined from Eq.

4. Thus, although the assumptions about the tortuosity certainly must add to the uncertainty,

they are apparently not influencing the result for ζ-potential.

3.4.3 Potential applications and implications

A plot of µobs vs µep (e.g. Fig. 3.4) relates an easily measured quantity, µep, to information about

how fast a molecule moves in the brain. This is useful in the context of experimental procedures

like iontophoresis. The µobs for a solute dictates whether it will move into or away from an

2 iontophoresis capillary. A neutral molecule, for example, has a µep of 0 m /Vs. By applying Eq.

-10 2 8, the µobs is expected to be 5.2 x 10 m /Vs. Therefore, in the OHSC and with the proviso that

the iontophoresis capillary ζ-potential matches the tissue ζ-potential, a neutral compound would

move toward the cathode. To eject a neutral compound iontophoretically, it must be in the anodic

chamber.

It is instructive to consider under what conditions EO flow contributes to the transport of

a neutral molecule by comparing the magnitudes of EO velocity and diffusive velocity. This is

done with the Peclet number, Pe. The Peclet number is the ratio of the time required to diffuse a

characteristic distance a, (~a2/D), to the time required to move with the fluid flowing at velocity

v over the same distance, a (a/v).

va Pe = (9) D

For a neutral solute, the Smoluchowski equation (Eq. 3) for ζ with the Stokes-Einstein equation

for D yields the following Peclet number.

36 6πraεζE Pe = (10) kT

In Eq. 10, r is the radius of the presumed spherical solute, k is Boltzmann’s constant, and T is temperature. If we assume that tortuosity influences diffusion and migration to the same degree, it does not appear in Pe. The field at which diffusive and electrokinetic velocities are equal, Ec, is

thus

kT E = (11) c 6πraεζ

Finally, equating Ec with a critical voltage difference over the characteristic length, ∆Vc/a, the

critical voltage is given as Eq. 12:

kT V = (12) c 6πrεζ

Note that the characteristic distance divides out. We can use a correlation165 to relate molecular

radius and molecular mass. For a 1 kDa (r = 0.84 nm) neutral molecule, the critical voltage is 19

mV (for the ζ-potential that we measured, -22 mV). For a 10 kDa neutral (r = 2.6 nm), the

critical voltage is only 6 mV. Clearly, this magnitude of potential difference is plausible, leading

to the conclusion that EO flow may in fact play a role in extracellular transport.

If EO flow does play a role, how would it influence charged neurotransmitters/

modulators? The direction of EO flow when ζ is negative is the same as the electrophoretic

direction of a cation. Thus, an electric field would drive a cation more rapidly, and an anion less

rapidly, in the presence of EO flow in comparison to the absence of EO flow. It is interesting to

note that the influence of dopamine, a cation, is felt outside the synapse. Further, its local

concentration and concentration dynamics are important in its actions.33, 166 The presence of an

electric field would cause dopamine to move relatively rapidly. In vertebrate brain, the anionic

37 amino acid glutamate is responsible for fast neurotransmission. Unrestricted diffusion of

glutamate over larger areas may cause synchronous activation of neurons and epilepsy. Thus this

type of transmission has to be tightly localized to the individual synapse from which glutamate is

released. The main system to regulate this process is reuptake. However, it is worth noting that

any electric field that exists would have very little impact on glutamate's motion- its

electrophoretic velocity and the extracellular fluid's EO velocity will counteract each other.

3.5 ACKNOWLEDGMENTS

The authors thank National Institutes of Health (R01 GM 44842) and the Swedish Research

Council/Medicine for grants that supported this work.

3.6 SUPPORTING MATERIAL

3.6.1 Results

Several other small dyes were also potential candidates for this study, but were excluded for the following reasons: Quantum yield too low: neutral red, tris(2,2’-bipyridine)osmium(II), tris(4,7- bis(4-sulfophenyl)-1,10-phenanthroline) ruthenium(II), bis(2,2’-bipyridine)(4,7-bis(4-sulfophenyl)-1,10-phenanthroline) ruthenium(II), bis(2,2’-bipyridine)(4-carboxy-2,2'-bipyridine) ruthenium(II). Atypical capillary electropherogram: Fluoresceinamine, isomer 1. Interacts with tissue: Rhodamine 6G, Rhodamine B.

38 3.6.2 Discussion

3.6.2.1 Analysis of the method

Most omitted dyes simply have a low quantum yield under our conditions. Fluoresceinamine,

isomer 1, on the other hand, has multiple peaks in the CE, indicating impurities. The rhodamine

dyes appear to associate with the cell surface or components of the extracellular matrix. Their µep

values appear reasonable according to their sizes and charges, but rhodamine 6G does not move

-10 2 in tissue at all, while rhodamine B has a µobs of -4.28 ± 0.86 x 10 m /Vs. The latter µobs is

smaller in the OHSC by a factor of 10 from what is expected from Eq. 8.

3.6.3 Materials and methods

3.6.3.1 Chemicals and solutions

Each of the following fluorescent dyes was diluted with HBSS, filtered, and frozen until use.

Rhodamine 6G, Rhodamine B, Neutral Red, tris(2,2’-bipyridine)osmium, and Fluoresceinamine,

isomer 1 were all obtained from Sigma. Three dyes, tris(4,7-bis(4-sulfophenyl)-1,10-

phenanthroline)ruthenium,167 bis(2,2’-bipyridine)(4,7-bis(4-sulfophenyl)-1,10-phenanthroline) ruthenium,168 and bis(2,2’-bipyridine)(4-carboxy-2,2'-bipyridine)ruthenium,169 were synthesized

in the laboratory using materials obtained from Sigma and previously reported procedures.(1-3)

3.6.3.2 Capillary electrophoresis

CE analyses of rhodamine 6G and rhodamine B were performed in a 40 μm inner diameter (ID),

360 μm OD capillary (118.4 cm total length, 26.8 cm from injection to detection) using -22.1

kV. CE of fluoresceinamine, isomer 1 used a 50 μm ID, 360 μm OD capillary (70.9 cm total

39 length, 48.1 cm from injection to detection) using -9.9 kV. These 3 dyes were detected at 495

nm. The remaining dyes listed above were not analyzed by CE.

Table 3.3. Dye concentrations

Dye Concentration (mM)

1 2.00 2 3.06 3 0.60 4 2.00 5 1.00 6 1.00 7 1.00 8 0.30

Table 3.4. CE parameters

Capillary Inner Total Lenth to Applied UV absorbance Fluorescent Compound Diameter (µm) Length (cm) Window (cm) Potential (kV) (nm) AlexaFluor 568 10 kDa‡ 50 50.3 31.5 -10.5 505 BODIPY FL 10 kDa‡ 50 50.3 31.5 -9.0 505 Fluorescein 70 kDa‡ 40 89.4 26.5 -22.0 505 Oregon Green 10 kDa‡ 40 87.4 25.5 -22.0 505 Rhodamine Green 3 kDa‡ 40 89.4 26.5 -22.0 505 Texas Red 70 kDa‡ 40 89.4 26.5 -22.0 595 Fluorescein sodium salt 40 118.4 26.8 -22.1 495 tris(2,2’-bipyridine)ruthenium 40 118.4 26.8 -22.1 495 Rhodamine 6G 40 118.4 26.8 -22.1 495 Rhodamine B 40 118.4 26.8 -22.1 495 Fluoresceinamine, isomer 1 50 70.9 48.1 -9.9 495 ‡Dextran Conjugate

40 4.0 DETERMINATION OF ζ-POTENTIAL AND TORTUOSITY IN RAT

ORGANOTYPIC HIPPOCAMPAL CULTURES FROM ELECTROOSMOTIC

VELOCITY MEASUREMENTS UNDER FEEDBACK CONTROL

Reproduced with permission from Analytical Chemistry 2009, 81 (8), 3001-3007.

ABSTRACT

Extracellular translational motion in the brain is generally considered to be governed by diffusion and tortuosity. However, the brain as a whole has a significant ζ-potential, thus translational motion is also governed by electrokinetic effects under a naturally occurring or applied electric field. We have previously measured ζ-potential and tortuosity in intact brain tissue, however the method was tedious. In this work, we use a four-electrode potentiostat to control the potential difference between two microreference electrodes in the tissue, creating a constant electric field. Additionally, some alterations have been made simplify our previous

procedure. The method entails simultaneously injecting two 70 kDa dextran conjugated

fluorophores into rat organotypic hippocampal cultures and observing their mobility using

fluorescence microscopy. We further present two methods of data analysis: regression and two-

probe analysis. Statistical comparisons are made between the previous and current methods as

well as between the two data analysis methods. In comparison to the previous method, the

41 current, simpler method with data analysis by regression gives statistically indistinguishable

mean values of ζ-potential and tortuosity, with a similar variability for ζ-potential, -21.3 ± 2.8 mV, and a larger variability for the tortuosity, 1.98 ± 0.12. On the other hand, we find that the current method combined with the two-probe analysis produces accurate and more precise

results, with a ζ-potential of -22.8 ± 0.8 mV and a tortuosity of 2.24 ± 0.10.

4.1 INTRODUCTION

Translational motion of molecules in the extracellular space of functioning tissues is typically

viewed as being governed by diffusion and tortuosity.18 Although not as well studied,

electroosmotic effects occur in tissues due to natural processes117 or experimentally applied fields.15, 17, 21, 126-128, 170 Electroosmotic flow is the bulk fluid flow created by an electric field in a

heterogeneous medium with a non-zero ζ-potential. In the brain for example, fixed charges on cell-surface functional groups and constituents of the extracellular matrix create a ζ-potential.78

The electroosmotic velocity is governed by the magnitude of the ζ-potential.

In order to determine the electroosmotic velocity in brain tissue, it is necessary to know

the ζ-potential. Some methods exist for determining the ζ-potentials of particulate objects (e.g., cells) and film-like objects (e.g., skin). ζ-potentials of particulate objects have been determined by electrophoretic techniques.139-145 Electroosmotic flow in thin samples, such as plant tissue can

be determined by flux through the tissue.146 Electroosmotic flow is described and quantitated as

an influential parameter in transdermal iontophoresis30, 131-136 and transdermal sampling (reverse

iontophoresis136, 138). Transport through individual skin pores can be observed by applying an

42 oscillating electric potential147 or by scanning electrochemical microscopy.30 The experimental arrangement for determining the electroosmotic velocity through pores, such as in skin or leaf epidermis, is simplified by the geometry of these biological film-like samples. A potential difference can be generated across the film-like sample. Electrokinetic phenomena then occurs perpendicular to the film-like sample, carrying fluid from one side of the film-like sample to the other.146 Quantifying ζ-potential and electroosmotic flow in a heterogeneous tissue matrix, on the

other hand, is technically challenging.

We have previously published a method of determining ζ-potential and tortuosity in

intact organotypic hippocampal slice cultures (OHSCs) using fluorescent probes.78 The hippocampus is of great interest because it is the center of spatial learning and memory.

Conditions such as Alzheimer’s disease and ischemia that occur in this region are particularly devastating. The experimental method places an OHSC on the stage of a fluorescence microscope. The velocities of several fluorescent probes in the tissue experiencing an applied electric field are measured. Relating the measured velocities, which are the sums of the electrophoretic and electroosmotic velocities, to the probes’ electrophoretic mobilities in free solution leads to a determination of ζ-potential and tortuosity. While successful, the method is not without drawbacks. The focus of this work is to simplify and reduce the effort required to make these measurements while decreasing or at least not increasing the variability in the previous method.

Potential drawbacks in the previous method include: 1) using agarose gel containing a

HEPES buffered salt solution (HBSS) as a contact between the electrodes and brain tissue. The contact between the gel and tissue is difficult to reproduce; 2) measuring, not controlling, the electric field. While the field is apparently constant (measured zone velocity independent of

43 time), it would be better to control the field; 3) using eight probes. Using eight probes was

helpful to establish experimentally the predicted linearity in a plot of observed vs. electrophoretic

mobilities. Now that the linearity is proven, there is no need to repeat the process. Consequently,

it may be possible to use fewer fluorescent probes; 4) using probes of various molecular weights.

Using probes of similar composition and molecular weight is preferred because tortuosity is

dependent on probe composition and molecular weight.109, 171, 172 In this work, the measured ζ-

potential and tortuosity are compared to our previous work to assess accuracy and precision.

Furthermore, we compare two data analysis approaches to obtain ζ-potential and tortuosity from

measurements using only two probes.

4.2 THEORY

We have previously published the general theory for determining ζ-potential and tortuosity.78

The experimental design relies on fluorescent probes with a range of electrophoretic mobilities.

In the work presented here, solutions of fluorescent probes are injected simultaneously in pairs

into an OHSC on the stage of an inverted microscope. Under the influence of an electric field

applied perpendicular to the optical axis, each fluorescent probe will move by electroosmosis,

electrophoresis, or both at an observed velocity, vobs, which is the product of the mobility, µ, and

the electric field, E. The electroosmotic mobility is called µeo. Equation 1 defines the relationship

22, 151 between µeo and ζ-potential, ζ.

εζ v = − E = μ E (1) eo η eo

44 Equation 1 shows that µeo is a function of ζ, the viscosity of the medium, η, and the permittivity, ε. The same relationship (with a change in sign) exists for electrophoretic velocity, vep, and mobility, µep. Thus, an observed mobility, µobs, is defined by equation 2 for an individual

molecule, where λ is the tortuosity78.

 εζ  vobs 1 1 μ obs = = ( 2 )(μ ep + μ eo ) = ( 2 )μ ep −  (2) E λ λ  η 

One method of analysis, implemented in our prior work, uses linear regression to

determine the parameters of interest. A linear regression of µobs on µep yields ζ and λ from the

intercept and slope using values of ε and η for water. Another method, which we will call the

‘two-probe’ method, takes advantage of the fact that we use a solution containing two electrophoretically and spectroscopically different fluorescent probes (1 and 2). Each probe moves independently at an observed mobility. The two probes’ observed mobilities, measured simultaneously in one experimental run, are used to calculate a single value of ζ and a single value of λ. Values of ζ and λ are determined as shown in equations 3 and 4. Here, the numerical

subscripts refer to the two probes.

(µ − µ ) ζ = µ ep1 ep2 − µ - ε  obs1 (µ − µ ) ep1  η obs1 obs2 (3)

µ − µ λ = ep1 ep2 e µ − µ obs1 obs2 (4)

Reported values for ζ and λ are obtained by averaging the results of the individual, two-probe

experiments.

45 4.3 EXPERIMENTAL SECTION

4.3.1 Chemicals and solutions

The culture medium was comprised of Basal medium Eagle (50%), Earl’s basal salt solution

(20%), horse serum (23%), penicillin/streptomycin (25 units/ml), L-glutamine (1 mM), and D-

(+)-glucose (7.5 g/L). All components were acquired from Sigma (St. Louis, MO) and used as received. The culture medium was warmed to 37 °C before use, but otherwise was kept in a refrigerator. D-(+)-glucose (27.5 mM) and MgSO4 (2.7 mM) were added to Gey’s Balanced Salt

solution (GBSS), all also purchased from Sigma, as were the following: HEPES-buffered salt solution (HBSS) containing (mM): 143.4 NaCl, 5 HEPES, 5.4 KCl, 1.2 MgSO4, 1.2 NaH2PO4,

2.0 CaCl2, and 10 D-(+)-glucose. HBSS was prepared with 18 MΩ purified water from a

Millipore Synthesis A10 system (Millipore, Billerica, MA), filtered and refrigerated at 2.6 °C. It

was warmed to 37 °C before use.

Fluorescent probes, Texas Red dextran conjugate 70 kDa (TR7) and Fluorescein dextran

conjugate 70 kDa (Fl7) were purchased from Invitrogen/Molecular Probes (Eugene, OR). TR7 and Fl7 were dissolved in HBSS to make 0.34 mM and 0.67 mM solutions, respectively, then filtered with 13 mm, 0.45 µm PTFE Millex filter units or equivalents (Millipore) and frozen until

use. The final solution containing both fluorophores had a concentration of 0.19 mM for TR7

and 0.29 mM for Fl7.

46 4.3.2 Organotypic hippocampal slice cultures

OHSCs were prepared according to the Stoppini culturing technique with small variations.84 The procedures below were authorized by the University of Pittsburgh IACUC. Seven-day postnatal

Sprague-Dawley albino rats were decapitated and the hippocampi dissected bilaterally. The hippocampi were cut perpendicular to the septo-temporal axis using a McIlwain tissue chopper

(The Mickle Laboratory Engineering, Surrey, England) into 500 µm thick slices. The OHSCs were placed on Millicell membrane inserts (Millipore CM), pore size 0.4 µm, over culture medium and incubated in 95% air/5% CO2 at 37 °C. The cultures are roughly ellipsoidal, with the major axis approximately running from the CA3 to the subiculum.

All OHSCs used for the data reported here were cultured for 3 to 10 days, when the tissue thickness is ~150 µm.84 Prior to an experiment, the culture medium was replaced with 37 °C

GBSS and incubated for 30 minutes. This exchange and incubation was repeated two more

times, where the final exchange was with 37 °C HBSS. The following manipulations took place

at room temperature. The insert membrane was cut using an X-Acto knife to separate the OHSC

and the membrane directly beneath it from the insert, as shown in Figure 4.1. To facilitate

handling of the tissue, a millimeter or two of membrane was left extending along the major axis,

and the combined membrane/culture was placed in the apparatus with the major axis parallel to

the applied field.

4.3.3 Four-electrode potentiostat

We have converted a three-electrode potentiostat (Princeton Applied Research Model 173,

Princeton, NJ) to a four electrode potentiostat using a differential amplifier in order to convert

47 the two measured reference electrode potentials into a single difference potential.173 This

potentiostat is capable of applying up to 100 V between the two ‘counter’ electrodes in our

apparatus (described below). To accommodate a high common mode voltage, a high

voltage operational amplifier, AD445 (shown as U1 and U2 in Supporting Information Figure S-

4.1) from Analogue Devices (Norwood, MA), with a power supply range of +/- 50 volts was used. The amplifier has a typical input bias current of +/- 1.4 pA and a typical input offset current of +/- 0.2 pA. These amplifiers were connected as buffer amplifiers and were the inputs to an AD629 differential amplifier (U3), also from Analog Devices. The amplifier that consists of U1, U2, and U3 allows a common mode range of +/- 42.5 V. Two reference electrodes were connected to the input ends of the amplifier. The reference electrode closest to the cathode was connected to the negative input end of the U3 differential amplifier by way of the buffer amplifier U1 (AD445). Another AD445, U2, was the buffer amplifier connected to the positive input end of U3 and to the other reference electrode. Buffer amplifiers U1 and U2 were connected to a +/- 45 V HP 6205C Dual DC Power (currently Agilent, Santa Clara, CA). A homemade +/- 15 V DC power supply supplied power to U3. Capacitors were used in the power supplies to remove noise, as shown on the bottom of Figure S-4.1. The output end of U3 was connected to the reference electrode input of the potentiostat. Platinum electrodes in each reservoir of the cell (described below) were the two counter electrodes. They were connected to the auxiliary and working inputs of the potentiostat. This configuration created a feedback loop such that the potential difference between the sensing regions of the two reference electrodes was constant.

48 4.3.4 Apparatus for tissue experiments

Illustrated schematically in Figure 4.1, the cell was manufactured locally using Lucite. The cell was comprised of a narrow, 44 mm long, rectangular channel separating two cylindrical reservoirs (19 mm in diameter and 12 mm deep). Each of the reservoirs contained a tubular platinum electrode (1 mm OD, 0.1 mm wall thickness, and ~3.8 cm long) obtained from

Goodfellow (Cambridge, England). The OHSC, already removed from the excess insert membrane as described in a previous section, was placed in the middle of the channel. Electrical contact between the OHSC and the platinum electrodes was made by filter paper, Grade 1

Whatman (Maidstone, England), soaked with HBSS. A piece of filter paper, with one side in the

HBSS reservoir, was laid just over one end of the tissue. The same was done with another strip of filter paper from the other reservoir to the other end of the OHSC, as shown schematically in

Figure 4.1. Two 100 µm (ID) fused-silica capillaries, separated by a fixed distance of 2.5 mm, were used as Luggin capillaries. They were each pulled to a tip of ~20 µm (OD) using a P-2000 capillary puller from Sutter (Novato, CA) and inserted into the ends of the OHSC facing each of the platinum electrodes. HBSS filled the Luggin capillaries and their associated vials. The vials each contained a Bioanalytical Systems Ag/AgCl reference electrode (3 M NaCl, West

Lafayette, IN). These reference electrodes were connected to the input ends of the differential amplifier as described previously.

Figure 4.1. The apparatus

The apparatus sits on an inverted fluorescence microscope for image analysis. Reservoirs on either side of the apparatus each contain a platinum counter electrode (labeled ‘-’ and ‘+’) and is half filled with HBSS. A

HBSS-saturated piece of filter paper spans from inside each reservoir to either end of the OHSC, which sits on a piece of membrane in the center of a channel. Pulled Luggin capillaries are inserted into the OHSC 2.5 mm apart. The distal ends of the capillaries are in vials containing Ag/AgCl electrodes filled with HBSS.

4.3.5 Injection capillary

Using the same capillary puller, a 250 µm (ID) fused-silica capillary was pulled to a tip of ~12

µm (OD). The capillary was lowered perpendicularly into the OHSC with a Sutter (Novato, CA)

MP-285 micromanipulator. The distal end of the injection capillary connected to a FemtoJet® express from Eppendorf (Hamburg, Germany), which was pressurized by a compressed nitrogen tank from Valley Natural Gas (Wheeling, WV). The FemtoJet® express was set to deliver an injection pressure of 200 to 300 hPa for 0.2 seconds, with a compensation pressure of 40 to 70 hPa.

50 4.3.6 Imaging

Imaging of the experiment was done with an inverted IX71 fluorescent microscope with a UPlan

Apo 4× objective lens, both Olympus products (Melville, NY), and a charge-coupled device camera (ORCA-285 Hamamatsu, Hamamatsu City, Japan). A DA/Fl/TA-3X-A Triple-band

“Pinkel” filter set from Semrock (Rochester, NY), with exciter 1 387 nm, exciter 2 494 nm, exciter 3 575 nm, triple-band dichroic mirror: 394–414 nm, 484–504 nm, 566–586 nm, emitter:

457, 530, 628 nm was used according to the fluorescent properties of the probes. SimplePCI 6.0 software from Compix (Cranberry, PA) was used for image acquisition and processing.

4.3.7 ζ-potential determination

Prior to the start of the ζ-potential measurements, an OHSC is injected with each fluorophore solution separately to set each exposure time using the autoexposure command for a two-color image on SimplePCI 6.0. In preparation for each measurement, a fresh OHSC was prepared and placed in the measuring cell on the stage of the microscope. The tip of the injection capillary was loaded with a solution containing both fluorophores at room temperature. It was lowered perpendicularly ~75 µm (approximately midway) into the new OHSC using the micromanipulator. We estimate that the FemtoJet® express delivered about 60 pL of the solution most of the time. Occasionally, mere placement of the capillary in the tissue delivered a larger quantity of the fluorophore solution through capillary effects. Once the injection was complete, the injection capillary was raised from the OHSC and the Luggin capillaries were lowered into the tissue. The time lag from the injection to the application of the electric field is long enough

51 that convective flow from the injection does not interfere with the electroosmotic flow during the measurement.

An electric field between 320 V/m and 2000 V/m was chosen arbitrarily for each run.

This range of values was limited on the low end by the need to measure accurately the motion of the fluorescent zones, and on the high end to avoid electroporation.159 The potentiostat’s applied voltage was determined based on the chosen field and the 2.5 mm reference electrode spacing.

Following application of the potential, one image was acquired every one, five, ten, or twenty seconds. The electric field was measured at the two Luggin capillaries by a multimeter to ensure that the system was controlling the field accurately. The same was done for the two platinum electrodes in the reservoirs.

Figure 4.2 shows still images taken at two times during an experiment. The Supporting

Information includes unenhanced and enhanced movies showing the fluorophore movement.

Positive movement of a fluorophore was defined as movement toward the cathode. Using the imaging software, we determined the velocities of the fluorescent zones as follows. The positions of the zones prior to the application of the field (t=0) are defined by the location of the maximum in fluorescence intensity. We then observed the movement of the zones under the influence of the field (using a software-generated video) to determine the point at which a well-defined zone was discernable from the initial zone, as a considerable amount of fluorescence typically remains at or near the position at t=0. The location at a particular time of the intensity maximum of the moving zone defined the zone velocity (N.B. – a multiple-point approach demonstrating constant velocity was used for 203 measurements in the previous publication78). The distance that the zones moved in and experiment varied from about 50 to several hundred µm.

Figure 4.2. Sample experimental run

The two images show both fluorophores injected into the CA1 of the OHSC. TR7 is red, and Fl7 is green. The top image is taken at time 0. The bottom image shows, at a time of 5 minutes, that the probes have moved towards the cathode. The bottom image’s contrast has been increased to increase the visibility of the fluorophores.

4.4 RESULTS

We measured fifty pairs of electroosmotic mobilities (one for each fluorescent probe) in OHSCs.

One pair’s derived values of ζ-potential and tortuosity (based on equations 3 and 4) were

obviously outliers (ζ = -132 mV and λ = 8.4). This pair was not included in the following analysis. We will first compare results (previous vs. current) for the directly observed quantity,

µobs. We then compare results (previous vs. current) for the derived parameters, ζ-potential and

53 tortuosity. Finally, we compare the two data treatments, regression vs. two-probe, using the

‘current’ data.

Table 4.1a compares the observed mobilities for each fluorescent probe from the previous

experimental method to those of the current method using a Student’s t-test. The number of runs is n; t is the Student’s t value; and p is the probability that the observed difference (t value) is due to chance alone, also known as the p-value. Table 1b shows the analogous variances, s2, the

degrees of freedom, abbreviated as df, as well as the F statistic, Fs, and its p-value, p. It is worth

restating that there are several differences between the previous method and the current method.

The statistical testing will determine whether the current, much simpler, method gives acceptably

equivalent results (t-test) with no poorer precision (F-test) than the previous method. The t-test in

Table 4.1a shows that for both Fl7 and TR7 the null hypothesis, that the two mobilities are the same, should be accepted. Additionally, the change in the method modestly improves the precision in both observed mobilities as shown by Table 4.1b with p-values of 0.07 for both fluorophores. Thus, the observed mobilities are equivalent to, and perhaps more precisely determined, by the simpler current method than the previous method. An increase in precision is helpful because it reduces the number of experiments required to establish a value of the ζ- potential with a given confidence interval.

54 Table 4.1a. Comparison of observed mobilities

Previous Method Current Method † † † † µobs SEM N µobs SEM n t p Fl7 1.85 0.24 22 1.76 0.13 49 0.40 0.69 TR7 3.66 0.71 12 4.12 0.26 49 0.88 0.38

Table 4.1b. Comparison of variances

Previous Method Current Method 2 ‡ 2 ‡ s df s df Fs P Fl7 1.32 21 0.78 48 1.69 0.07 TR7 6.10 11 3.24 48 1.88 0.07

†Units: 10-9 m2/Vs ‡Units: 10-18 m4/V2s2

A linear regression of µobs vs. µep from the current method (n=98) yields a slope of 0.26 ±

0.03 and a y-intercept of 4.3 ± 0.2 × 10-9 m2/Vs. The stated errors are the SEMs. The slope is

equal to the inverse square of the average tortuosity. Therefore, the average tortuosity is 1.98 ±

0.12 (see Table 4.2a). Inserting the intercept into equation 2 yields an average ζ-potential of -

21.3 ± 2.8 mV. Table 4.2a shows two-tailed t-test comparisons of ζ-potential and tortuosity

determined with the current method vs. the values from the previous method.78 With

probabilities of 0.84 for ζ-potential and 0.26 for tortuosity, we infer statistically indistinguishable

values. F-test comparisons, as shown in Table 4.2b (p-values of 0.60 for ζ-potential and 5.2 × 10-

5 for tortuosity), show the variability in ζ-potential is indistinguishable from the previous

method’s. However, this table shows we have increased the variance in the tortuosity determined

by the current method using linear regression for data analysis in comparison to the previous

method (which also used linear regression).

55 Table 4.2a. Comparison of ζ-potential and tortuosity of the two experimental methods

Previous Method Current Method Mean* SEM* N Mean* SEM* n t p ζ -22 2 203 -21.3 2.8 98 0.20 0.84 λ 1.83 0.06 203 1.98 0.12 98 1.1 0.26

Table 4.2b. Comparison of variances

Previous Method Current Method 2 § 2 § s df s df Fs p ζ 812 201 773.82 96 0.95 0.60 λ 0.73 201 1.42 96 1.93 5. 2 × 10-5

*Units: for ζ, mV; for λ, unitless §Units: for ζ, mV2; for λ, unitless We now turn to a comparison of data analysis methods. This comparison is based only on data from the current method. In the regression analysis, all 98 data are used to find ζ-potential and tortuosity. In the two-probe analysis, 49 values of ζ-potential and tortuosity are averaged.

Table 4.3a displays the results of the two analysis methods, namely linear regression and the two-probe method. The two analysis methods yield statistically indistinguishable values for both

ζ-potential and tortuosity, as shown by their p-values of 0.60 and 0.24, respectively. However, the F-tests in Table 4.3b show that the precision greatly improves with the use of the two-probe analysis.

56 Table 4.3a. Comparison of ζ-potential and tortuosity of the two analysis methods

Regression Analysis 2-Probe Analysis Mean* SEM* N Mean* SEM* n t p ζ -21.3 2.8 98 -22.8 0.8 49 0.52 0.60 λ 1.98 0.12 98 2.24 0.10 49 1.18 0.24

Table 4.3b. Comparison of variances

Regression Analysis 2-Probe Analysis 2 § 2 § s df s df Fs p ζ 773.82 96 29.48 48 26.2 < 0.0001 λ 1.42 96 0.52 48 2.74 0.0001

*Units: for ζ, mV; for λ, unitless §Units: for ζ, mV2; for λ, unitless

4.5 DISCUSSION

4.5.1 Comparison of methods

We made several alterations to the original experimental setup and procedure to simplify and improve the method of determining ζ-potential and tortuosity in intact OHSCs. One such alteration is the use of HBSS in conjunction with filter paper instead of agarose gel slabs. This eliminates the variability in the contact between the tissue and the agarose and eliminates the potential for variability among agarose compositions. Furthermore, the fluid flows more easily within the filter paper than in the agarose gel. This lower resistance to flow minimizes possible pressure differences. Originally, Ag/AgCl electrodes were inserted into the tissue cultures in the experimental region to measure the electric field during the experiment.78 These electrodes, although small, were invasive and cumbersome. By replacing them with Luggin capillaries,

57 pulled to a tip, the experimental manipulations become easier. The pulled tips of the Luggin

capillaries are smaller (~20 µm OD) than the diameters of the Ag/AgCl electrodes (~100 µm

each). Also, the capillaries allow the use of larger, more stable, commercially available (BAS)

Ag/AgCl electrodes placed in remote vials. By using larger electrodes, amplifiers demanding

larger currents are useable, thus simplifying potentiostat design and decreasing drift due to

changes in the reference electrode potential occasioned by the operational amplifier current

requirement. Most importantly, these electrodes no longer measure the electric field throughout

the experiment, but participate in controlling the field between the capillaries with the use of a

four-electrode potentiostat. Controlling the field in this manner reduces the variability in the field

throughout the experiment. Lastly, the capillaries are now spaced further apart within the tissue culture, so as not to interfere with the fluorophore migration.

The four-electrode potentiostat replaces the use of a voltage supply and reference

electrodes for measurement. A three-electrode potentiostat controls the potential between a

working electrode and a reference electrode by applying current through an auxiliary electrode.

Coupling a differential amplifier to the reference electrode permits the use of two reference

electrodes173 allowing for the control of a potential difference without the requirement that one of

the electrode potentials is zero (ground). In this configuration, the potentiostat controls the

potential between the two reference electrodes in a feedback loop. Each reference electrode’s

sensing point is defined by a Luggin capillary. The Luggin tips are small and do not interfere

with the experiment. They are inserted along the major axis of the tissue such that the electric

field within the tissue is maintained by current between the counter electrodes. The potentiostat

maintains the potential difference between the counter electrodes in a range of 14 to 46 V

(depending on the tissue and the field that we are trying to achieve). Therefore, the potential at

58 the reference electrodes can be considerably larger than about 15 V vs. ground. This necessitated

the use of amplifiers that can manage a high common mode voltage.

The hippocampus is a highly structured region. There are three areas that contain

neuronal cell bodies, and thus show up clearly in images of sections of the hippocampal region,

namely, the CA1, CA3, and the dentate gyrus. From the previous work, we know that there are

small differences in ζ-potential and tortuosity among these three regions.78 Of the 98

measurements in the data set for the current experiments, 26 are in the CA1, 36 are in the CA3,

and 36 are in the dentate gyrus. A single-factor (regions) analysis of ζ-potential shows a p-value of 0.1 indicating borderline statistically insignificant differences among the regions. In this paper, we have chosen to ignore those (small) differences, thus including observations from the all the regions in a single data set. As Table 3.1 shows, the current method yields equivalent results for the experimentally observed quantity (in comparison to the previous method) and it is possibly more precise (p = 0.07).

78 Since the linearity of the relationship between µobs and µep is established for OHSCs, it

is simpler to use only two probes for the regression analysis rather than eight probes. The

potential disadvantage of the two particular probes used here is their relatively similar

electrophoretic mobilities. The electrophoretic mobility range in the previous method is almost

six times larger than the range in the current method. The range of the experimentally observed

mobilities is similarly reduced by about a factor of six. The smaller range in the current method creates a greater sensitivity to additive random errors in the observed mobility than in the previous method. We are thus concerned about suffering an increase in variability of the ζ-

potential and tortuosity because of the lower mobility range. Although two-tailed t-tests of both

ζ-potential and tortuosity show that the current method (using only two probes) yields

59 statistically indistinguishable results from the previous method (see Table 4.2a), the F-test in

Table 4.2b test reveals that the alterations increase the variability of tortuosity. The variability in

ζ-potential, however, does not increase. The result that there is no observable change in the error of the ζ-potential is favorable, but the increased variance in the tortuosity is disappointing.

Fortunately, the two-probe method of analysis leads to improved precision as discussed next.

According to the two-tailed t-tests in Table 4.3a, the two-probe analysis is a viable analysis method. The significantly better precision of the two-probe analysis method is a result of reduced variability probably due to differences in tissue cultures. The two-probe procedure measures two observed mobilities in the same conditions, since they are injected simultaneously into the same location on the same tissue culture. The hippocampal formation is a highly structured entity in which there are variations in composition (cell bodies vs. projections) over small (~100 µm) distances. While we attempt to inject the probes into the cell-body-containing parts of the tissue, there is certainly a good chance for variability in this aspect of the measurement. Thus, it is not entirely unexpected to find that the variability in the measurement is reduced with the two-probe approach for data analysis.

4.5.2 Accuracy of the method

As Figure 4.2 and the video clips in the Supporting Information show, the fluorescent probes not only move linearly in the direction of the electric field, but also the initially well-defined zone disperses as it passes through the tissue. It has been shown by computations relating to capillary electrochromatography that in a packed bed, even with a constant ζ-potential, there is a distribution of velocities because of the locally varying electric field.174 Over the several minute

60 experimental time, the root mean square diffusion distance based on known diffusion

coefficients106, 171 and tortuosity is approximately 40 – 60 µm. As the observed dispersion is

greater than this value, it is unlikely that diffusion plays an important role in the observed

processes, at least in the two dimensions defined implicitly in Figure 4.2. On the other hand, we

have observed earlier that the fluorescent probes can reside in the insert membrane below the

cultured tissue.78 This population of probe molecules in the membrane moves very slowly in the

electric field. Consequently, the injection site is always visible, as is a tail of fluorophore behind

the zone moving in the tissue. The tail arises, we think, from probe diffusing out of the

membrane into the tissue and vice-versa.

The only extant value of ζ-potential is the value from the previous method. To the degree

that the previous method is accurate, the current method is accurate. Tortuosity has been more

widely investigated. The two dextran-conjugated fluorophores, Fl7 and TR7, were chosen and

paired because they have the same molecular weight, 70 kDa. Previous research has shown

correlations between probe molecular weight, or more properly, hydrodynamic radius, and

tortuosity.78, 109-111 Using probes with the same molecular weight decreases error in tortuosity

resulting from variations in probe radii.171 As far as we are aware, there are no measurements of

tortuosity in OHSC other that the value from the previous method. However, there are values for

70 kDa fluorescent dextran in the cortex of rat. The value of the tortuosity in acute rat cortical

slices is 2.25171. The value in cortex in vivo is 2.68109. Values of tortuosity for 3 kDa fluorescent dextran in acute hippocampal slices are lower than comparable values in cortex and neocortex.18

While there are not enough data for a firm conclusion, it appears that measurements of tortuosity

by this method are accurate.

61 Such measurements could be extendable to in vivo conditions. However, several

challenges arise in an in vivo experiment. The most significant is imaging of a fluorophore that is

no longer in an optically thin sample. ’Seeing’ the region of interest is essential. Consequently, many in vivo diffusion measurements are in the easily accessible cortex.18 Multiphoton

experiments,175 with currently available fluorescent probes, or the use of near IR probes176 may

be helpful. Injections of the fluorophore remain in the tissue following the conclusion of the

experiment. Therefore, the fluorophore must be non-toxic to the animal. In addition, electrodes

must be placed appropriately for electroosmotic flow to occur. This would be especially

challenging for measurements in the interior of the brain. Current neurosurgical methods use

transparent conduits,177 for example, for endoscopic visualization and manipulation with

corresponding instruments to access inner brain regions. Similar arrangements can be imagined

for an in vivo ζ-potential and tortuosity determination. By slightly changing the approach to

using contrast agents instead of fluorescent dextran conjugates, magnetic resonance imaging

(MRI) and computed tomography (CT) scans can visualize probe movement without a full

craniotomy. Electrodes and the injection capillary would still need to be inserted, but this can be

minimally invasive.

In conclusion, the new experimental method reduces the variability of the experimentally

measured mobilities. Moreover, the two-probe analysis method is accurate and more precise in

determining ζ-potential and tortuosity, while also being simpler than the previous method.

ACKNOWLEDGMENT

The authors gratefully acknowledge funding from the National Institutes of Health (R01 GM

44842) and the Swedish Research Council/Medicine.

62 4.6 SUPPORTING INFORMATION

ABSTRACT

The supporting information section contains Figure S-4.1, the circuit diagram of the differential

amplifier in the four-electrode potentiostat, and two video clips. Video S.2 is a contrast enhanced

version of Video S.1. Both movies show three injections: in the CA1, CA3, and dentate gyrus

regions. An electric field of 1.4 V/mm is applied across the tissue. Images are acquired every 20

seconds. During the course of 5 minutes, as shown by the time stamp in the upper left corner,

TR7 (in red) and Fl7 (in green) move according to their observed mobilities. TR7 separates from Fl7 as it moves towards the cathode at a greater velocity than Fl7.

Figure S- 4.1. Circuit diagram of the differential amplifier in the four-electrode potentiostat

64 5.0 IONTOPHORESIS INTO A POROUS MEDIUM DEPENDS ON THE ZETA

POTENTIAL OF THE MEDIUM

ABSTRACT

Iontophoresis uses electrically facilitated transport to deliver solutes into living tissue. Often,

iontophoretic ejections into brain tissue are confined to millisecond pulses for highly localized

delivery. This paper investigates solute penetration using extended iontophoretic ejections into

brain tissue. Furthermore, the effect of the ζ-potential on ejections is examined using poly(acrylamide-co-acrylic acid) hydrogels with various magnitudes of ζ-potentials, including

that similar to hippocampal brain tissue. Parameters that effect solute penetration distance

include the magnitude of the applied current, solute properties, and the ζ-potential of the matrix.

A theoretical calculation is presented as a quantitative method of predicting solute ejection

profiles using the measured properties of the system. The theory eliminates the need for

calibration of each iontophoretic experiment. Experiments also show that with an appreciable

matrix ζ-potential, such as that of brain tissue, solute transport may be directed to specific

locations following the direction of current flow.

65 5.1 INTRODUCTION

Iontophoresis is a popular method of solute delivery that uses the application of an electric field

via constant current to deliver solutes to tissue. Electrokinetic processes motivate solutes to be

transported into tissue. In addition to delivering solutes directly from a pipette into the brain,

Lerner et al. use iontophoresis to deliver compounds to the brain through the nasal mucosa.21

Brain iontophoresis is also used to stimulate neuronal pathways128 and understand the brain’s

physicochemical properties and transport mechanisms.128 Outside of brain delivery,

iontophoresis is used to deliver solutes transdermally, transocularly,178, 179 and transurethrally.180

Iontophoresis into the brain has the advantage of localized delivery that circumvents the

blood-brain barrier. The main drawback to iontophoresis is its unpredictability. Historically, the

transport number was used to correlate the number of moles of ejected ionic solute with the

applied current.181, 182 The correlation between the transport number and the ejected solute

concentration is inadequate.170 In practice, there is significant variability from one iontophoretic

ejection pipette to another. Previous attempts to quantify iontophoretic ejections have had limited

success.183-185 Researchers have used ion-selective microelectrodes,18, 186-188 carbon fiber

microelectrodes,189-191 and fluorescence microscopy192 to measure solute ejection outside of the

ejection pipette. Today, each ejection pipette must be pre-conditioned and calibrated in order to

obtain a predictable distribution profile outside of the pipette tip.170, 191, 193

Trubatch and Van Harreveld proposed that electrokinetic processes within the tissue may

account for some of the lack of agreement between predicted and experimental ejection

profiles.194 The theory was quickly refuted by Norman and others, stating that pulsed iontophoretic currents are low (≤100 nA) and ejections are too short to create appreciable

electrokinetic contributions within the tissue.193, 195-197 Until recently, the standard view was that

66 iontophoresis into the brain can only convey charged solutes into tissue via charge migration in

the iontophoretic pipette and diffusion thereafter into the tissue.120, 193, 195-198

Recently, Herr et al. demonstrated the significance of electrokinetic flow in the pipette during iontophoretic ejections into buffer solution using electrochemical detection adjacent to the pipette tip.198 Applying an electric field within the pipette creates an electrokinetic bulk fluid

flow known as electroosmotic flow, EOF.22 EOF arises as a result of an electric field’s

motivating force on the mobile counterions to the fixed charges on the pipette walls. A solute in

an electric field moves according to its electrophoretic velocity and the fluid’s electroosmotic

velocity in addition to diffusion. Therefore, Herr et al. were able to demonstrate ejections of

neutral solutes as a result of the EOF in the pipette.191, 198 The magnitude of EOF is proportional

to the ζ-potential. The ζ-potential is the potential at the shear plane between the moving

electrolyte solution and the immobile counterions along the surface. The magnitude of ζ is

governed by the surface charge density and the electrolyte concentration. Skin tissue, for

example, has an appreciable ζ-potential within pores and a low ζ-potential elsewhere.30, 135, 149, 199

Incorporating EOF into transport models leads to better predictions of solute delivery in

transdermal iontophoresis.30, 199 EOF is used in tumor dewatering, for example, where an electric

field is used to direct the migration of water out of tumors.137 Furthermore, Xu et al. were able to show sustained EOF in brain tissue during extracellular electroosmotic sampling.93 However,

EOF is not addressed in brain iontophoretic applications.

We have recently reported that brain tissue cultures have a ζ-potential of approximately -

22 mV.77, 78 We propose that a solute can be transported by virtue of the significant ζ-potential of brain tissue and the resulting EOF. Such transport will most easily be seen by prolonging iontophoretic ejections, as concentration gradients, and thus transport by diffusion, is fast on

67 short timescales. It is possible that EOF plays a part under conditions such as Nicholson’s

iontophoretic ejections using 60 – 100 nA for 50 second time intervals196 and in Herr’s

iontophoretic ejections periods of 30 to 60 seconds191.

5.1.1 Theory

The degree to which electrokinetic processes take part in iontophoresis is can be assessed by knowing the Peclet number, Pe. Pe, as shown in Equation 1, is a dimensionless ratio of the time a solute requires to diffuse a characteristic distance, ā, and the time a solute requires to move the

same distance with a velocity, v . D is the solute diffusion coefficient in solution.

va Pe = (1) D

Solute velocity, v, contains three components: an electrophoretic velocity, vep, an electroosmotic velocity, veo, and a pressure-induced velocity, vp. We assume that vp is small compared to vep and

veo, and therefore is not accounted for in this derivation (refer to SI for further information). Each

of the remaining two velocities is governed by the magnitude of the electric field. Equation 2 relates the electric field, E, to the current applied to the system, i:

E = jρ = i ρ (2) E A E

where j is the current density, ρΕ is the resistivity of the fluid, and A is the cross-sectional area.

Equations 3 and 4 define electrophoretic and electroosmotic velocities as products of electric field and the corresponding mobility, µ.

vep = µep E (3)

− εζ − εζ iρ v =  E = µ E =   E (4) eo  η  eo  η  A

68 Equation 4 shows that the electroosmotic velocity is a function of the ζ-potential of the matrix

(ζ), and the permittivity (ε) and viscosity (η) of the fluid. Incorporating Equations 2 through 4

into Equation 1 yields a theoretical calculation of Pe, PeT, in Equation 5 based on measurable properties.

iρE a  εζ  Pe = µ − (5) T A D  ep η 

The concentration profile of a solute being transported into tissue by iontophoresis in the

case where µep and µeo are both zero is governed only by diffusion. For the case in which the

iontophoretic current is on long enough to reach steady-state,191 the concentration profile takes

on the familiar shape described by Equation 6.200 Here, we have made the assumption that the

solute emanates from a hemispherical source. In Equation 6, the concentration, C, is a function

of the concentration at the surface of the hypothetical hemisphere of radius, ā, the maximum

concentration, C0, and the distance from the hemispherical surface, r.

C=C 0 r (6) ā In the case where there is convection as well as diffusion, we can solve for the steady-state

condition in the hemispherical coordinate system as follows. Equation 7 describes solute flux:

= C C (7)

where퐽 푣 J−, and퐷∇ v are the flux and velocity of the solute, respectively. The first term in the equation

refers to the flux due to convection, which includes both veo and vep. The second term relates to

the diffusional flux resulting from a concentration gradient. We assume the electric field

distribution from the pipette tip is equal to the field distribution from a hemispherical electrode,

(r) = i.e., 2 where E0 is the field at r = ā, leading to Equation 8. ā 2 0 퐸 r 퐸 C(r) = C(r) (8) obs 2 r ā 푑 2 0 퐽 µ r 퐸 − 퐷 푑 69 The observed mobility, µobs, is the sum of µeo and µep. Note that in the case of no electric field, the electrokinetic mobility is zero and the flux is governed solely by the diffusion of the solute.

Steady-state is achieved in a spherical coordinate system when dr2J/dr = 0. Thus, Equation 9 describes the steady-state condition.

C C C 2 2r r = 0 (9) r r 2r 푑 푑 2 푑 2 휇표푏푠퐸0ā 푑 − 퐷 푑 − 퐷 푑 = (10) 푃푒 2 ′ 2 휌 − 휌� 퐶 퐶′′ Equation 10 is Equation 9 with the derivatives with respect to radius indicated as primed quantities, the dimensionless distance, ρ, is r/ā, and Pe is the based on the electric field at the tip

(ρ = 1), i.e,

a Pe = (11) 휇obs퐸0 This equation퐷 can be solved by solving the first order differential equation for the function F =

C’, and then solving the resulting equation for C’ by integration with boundary conditions:

Boundary Condition 1: C( ) = 0

Boundary Condition 2: C(휌 → ∞1) = C0

휌 → C( ) = C PeE (12) − �휌 1−e −Pe 휌 0 � 1−e E � where PeE is the experimental Peclet number. When the electrokinetic effects are negligible, PeE tends to zero. Expanding the exponentials as Taylor series and truncation of quadratic and higher order terms leads to the recovery of Equation 6.

We hypothesize that a larger magnitude tissue ζ-potential will be associated with larger ejection distance in a given time period. As a tissue’s ζ-potential cannot be altered, we will test this concept, using synthetic poly(acrylamide-co-acrylic acid) hydrogels are used as tissue

70 models with varying ζ-potentials. We examine the degree of fluorescent dextran penetration as a

function of matrix ζ-potential and applied current. We also show that prolonged ejections do not

create cell death in brain tissue as compared to controls. Finally, we use tris(2,2’-

2+ bipyridine)ruthenium (Ru(bpy)3 ) as a representative cationic solute for which electrophoresis

and electroosmosis are significant and in the same direction. By understanding the EOF in both

the delivery system and in the tissue itself, we can estimate the ejected solute distribution.198

5.2 EXPERIMENTAL

5.2.1 Chemicals and Solutions

Unless otherwise noted, the following materials were purchased from Sigma (St. Louis, MO) and used as received. Solutions were prepared with Millipore Synthesis A10 system 18 MΩ purified water (Millipore, Billerica, MA). Glucose-free HEPES-buffered salt solution (G-f HBSS) contained in mM: 143.4 NaCl, 5 HEPES, 5.4 KCl, 1.2 MgSO4, 1.2 NaH2PO4, and 2.0 CaCl2. G-f

HBSS was filtered, stored frozen, warmed to room temperature, and ultrasonicated for ten

minutes before use. HBSS contained the same quantities as G-f HBSS with an additional 10 mM

D-(+)-glucose and underwent the same preparation and storage process. GBSS was made up of

27.5 mM D-(+)-glucose and filtered 2.7 mM MgSO4 supplemented Gey’s Balanced Salt solution. GBSS was stored in the refrigerator and warmed to 37 °C before use.

Culture medium contained the following from Gibco (Invitrogen, Eugene, OR): 50 %

Opti-MEM, 25 % heat-inactivated Horse Serum, and 25 % Hanks’ Balanced Salt Solution,

71 supplemented with 2 % B-27 vitamin and 1 % D-(+)-glucose from Sigma.201 Medium was filtered, refrigerated for storage, and warmed to 37 °C before use.

5.2.2 Fluorophores

The following dextran conjugates were obtained from Invitrogen. Texas Red dextran conjugate

70 kDa (TR70) and Texas Red dextran conjugate 3 kDa (TR3) were diluted to 0.06 mM and 0.2

mM, respectively with G-f HBSS, filtered, and frozen until use in the current ejection

2+ experiment. Tris(2,2’-bipyridine)ruthenium (abbreviated Ru(bpy)3 ) from Sigma was diluted to

1.3 mM in 150 mM NaCl supplemented with 5 mM HEPES, pH 7.4 in the high electrolyte

solution, and diluted to 1.3 mM in 5 mM NaCl supplemented with 5 mM HEPES, pH 7.4 in the

low electrolyte solution.

A TR70/F70 mixture was made in G-f HBSS with the final concentration of 0.19 mM

TR70 and 0.29 mM F70 for use in ζ-potential measurements. For diffusion coefficient measurements, the following fluorophores were dissolved in G-f HBSS to final concentrations of:

2+ TR70 (0.03 mM), F70 (0.03 mM), TR3 (0.67 mM), and Ru(bpy)3 (3.51 mM). A stock solution of 0.35 mM propidium iodide in GBSS was frozen until use for cell death analysis.

5.2.3 Determination of Diffusion Coefficients

Diffusion coefficients were determined as described in Beisler et al.202 G-f HBSS mobile phase pumped through the system using a Pico Plus syringe pump (Harvard Apparatus, Holliston, MA).

The sample solutions were injected using an HP 1050 autosampler (Hewlett-Packard, Palo Alto,

CA) into a 1 μL loop in a VICI 6-port Cheminert Injector (Houston, TX). The detector from an

72 ISCO 3850 Capillary Electropherograph (Teledyne ISCO, Lincoln, NE) was used for UV detection (215 nm) of the fluorophores. Signals were collected by Peaksimple 3.29 (SRI Inc.,

Torrance, CA) for analysis.

Data were imported into Origin 7.5 (Origin Lab Cooperation, Northampton, MA) for differentiation, followed by the determination of the first and second central moments using

PeakFit version 4 (Systac Software Inc., San Jose, CA). Diffusion coefficients were calculated from the slopes of second moment versus first moment linear plots. In addition, errors for the diffusion coefficients were calculated based on the error of the slope, assuming negligible error in the flow rate, capillary length and capillary diameter.

5.2.4 Electrophoretic Mobility Determination by Capillary Zone Electrophoresis (CZE)

The electrophoretic mobilities of the fluorophores in G-f HBSS were determined via CZE with diode array UV absorbance detection using 50 µm inner diameter (ID) capillaries and an Agilent

CE system (Agilent Technologies, Palo Alto, Ca). Hydrodynamic injection (25 mbar, 4 second)

was employed, and all experiments were performed at 25°C. For TR70 (λdetn = 595 nm), TR3 (595

2+ nm), F70 (494 nm), and Ru(bpy)3 (452 nm) an unmodified fused silica capillary (Polymicro

Technologies, Phoenix, AZ) was employed. The capillary was preconditioned with 1 M NaOH for 15 minutes, 18 MΩ-cm water for 15 minutes and G-f HBSS for 15 minutes, and was flushed with running buffer for 2 minutes in between injections. Rhodamine B (544 nm) was used as a neutral flow marker. The capillary was 33.0 cm in total length, 8.5 cm to the detector, and the separation voltage was 6.0 kV (182 V/cm). Fluorophores were dissolved in G-f HBSS and had concentrations ranging from 0.3 to 1.7 mM.

73 5.2.5 Synthesis of Hydrogels

Poly(acrylamide-co-acrylic acid) hydrogels were prepared with the purpose of creating matrices

with different ζ-potentials and tortuosities similar to those of brain tissue18 with a thermally-

initiated radical polymerization reaction. The total weight of acrylic acid, acrylamide, and

bisacrylamide was kept constant (279 mg/5.00 mL) while the w/w ratios of acrylic acid and

acrylamide were of 25 %, 10 % and 0 %. The bisacrylamide cross-linker was maintained at a

fixed weight percentage of approximately 1.4 %.

Table 5.1. Hydrogel compositions

25% 10% 0% weight (mg) mmol weight (mg) mmol weight (mg) mmol Acrylamide 206.0 2.90 247.5 3.48 275.0 3.87 Acrylic Acid 69.0 0.96 27.5 0.38 0.0 0.00 Bisacrylamide 4.0 0.03 4.0 0.03 4.0 0.03

The reagents were weighed and dissolved in 4.5 mL of ultrapure water (Cayman

Chemical, Ann Arbor, MI) and stirred vigorously for 5 minutes. The solutions were then titrated to a pH range of 7 to 8 using an aqueous 1 M NaOH solution. The final solution volumes were brought to 5.0 mL with ultrapure water. The polyacrylamide solutions were then deaerated with nitrogen for 20 minutes at room temperature. Immediately following, 5.0 mg of APS (0.02 mmol) and 5.2 μL of TEMED (0.03 mmol) were quickly added to each solution, which was then vigorously stirred for another 3 minutes at room temperature. Each clear, colorless solution was poured into a 50 mm x 16 mm Pyrex glass dish and placed inside an oven at 100 ± 5 °C for 2 hours. The Pyrex dishes were removed from the oven and allowed to cool for 10 minutes at room temperature. Using a glass pipette tip, the colorless and clear hydrogels were separated from the

Pyrex dish walls and slowly peeled from the bottom of the dish. The hydrogels were

74 independently placed in plastic dishes filled with approximately 25 mL G-f HBSS solution and shaken lightly at 10 °C overnight. The buffer solution was exchanged the following day with 25 mL of fresh G-f HBSS, placed back in the shaker at 10 °C for another 24 hours. Finally, the G-f

HBSS was exchanged once more (25 mL), and the hydrogels were placed in the refrigerator where they were allowed to equilibrate. A final pH check was performed to assure the hydrogels were pH

7.40 with full immersion of the probe into the hydrogel before further characterization.

5.2.6 ζ-Potential and Tortuosity

The experimental apparatus was used without modification from our previously reported setup.77

Briefly, a four-electrode potentiostat coupled with a differential amplifier and a pair of reference electrodes with Luggin capillaries was placed in the hydrogel. G-f HBSS was used as the electrolyte solution in the reservoirs. An Olympus IX71 inverted fluorescence microscope

(Melville, NY) imaged the experiment with an Olympus UPlan Apo 4× objective lens and a charge-coupled device camera (ORCA-285 Hamamatsu, Hamamatsu City, Japan). A DA/Fl/TA-

3X-A triple-band “Pinkel” filter set (Semrock, Rochester, NY), with exciter 1 at 387 nm, exciter 2 at 494 nm, exciter 3 at 575 nm, triple-band dichroic mirror: 394–414 nm, 484–504 nm, 566–586 nm, emitter: 457, 530, 628 nm was used. Image acquisition and post-processing was completed using SimplePCI 6.0 software from Compix (Cranberry, PA).

In order to set the exposure times, a piece of hydrogel was injected with each fluorophore solution separately. A fresh piece of hydrogel was cut to size and placed in the apparatus channel on the microscope stage. Using a P-2000 capillary puller from Sutter (Novato, CA), a 150 m ID fused-silica capillary was pulled to a tip opening of approximately 12 m to create an injectionμ capillary. The injection capillary was filled with the TR70/F70 fluorophoreμ mixture and lowered

75 into the hydrogel using a Sutter MP-285 micromanipulator to a depth of half the hydrogel’s thickness. The injection capillary was connected to the FemtoJet® express (Hamburg, Germany) to deliver a pressure injection via a compressed nitrogen tank (Valley Natural Gas, Wheeling,

WV). The injection pressure was set to 200 to 250 hPa for 0.2 to 0.5 seconds and 0 to 20 hPa compensation pressure. The exact parameters were varied within these ranges so that the volume of ejected fluorophore was appropriate for the sensitivity and dynamic range of the CCD camera.

Following fluorophore ejection, the injection capillary was removed from the hydrogel. The

Luggin capillaries were lowered to a depth of half the hydrogel’s thickness on either side of the injection spot. This process was completed an average of two times per hydrogel piece, and all experiments were conducted at ambient temperature.

An electric field between 60 V/m and 250 V/m was applied to each run. One image was acquired every second. The electric field was monitored at the two Luggin capillaries and at the platinum electrodes in the reservoirs by two multimeters to ensure accurate control of the electric field. The velocities of the TR70 and F70 fluorophores were measured in post-processing.

Fluorophore positions were recorded for 6 to 9 timepoints (30 to 60 seconds apart) per experimental run to arrive at an average velocity. The ζ-potential and tortuosity for each hyrogel were calculated from the average velocities as previously described.77

5.2.7 Porosity of Hydrogels

The water content of each hydrogel was determined by dehydration and the mass of water lost. A

10 mm x 10 mm section of hydrogel was cut and placed inside of a pre-weighed 1.8 mL ROBO autosampler vial (VWR, West Chester, PA). The aggregate weight was recorded for each vial, and placed in an oven at 110 °C for several hours until completely dehydrated. The vial was removed

76 from the oven and reweighed. The water content, by weight, was determined by subtracting the post-dehydration weight from the pre-dehydration weight. A minimum of three sections of each type of hydrogel was completed to give with a final porosity data provided as an average with standard error of the mean.

5.2.8 Conductivity of Hydrogels

Due to the high porosity of the hydrogels, the conductivity was assumed to be that of G-f HBSS.

The conductivity of G-f HBSS was measured using a CON 6/TDS 6 conductivity/TDS meter

(Eutech Instruments Pte Ltd/Oakton Instruments, Vernon Hills, IL).

5.2.9 Organotypic Hippocampal Slice Cultures (OHSCs)

The organotypic hippocampal slice culture (OHSC) method, developed by Stoppini et al. was used with slight variations.84 The following procedures were approved by the University of

Pittsburgh IACUC. Bilateral dissections of the hippocampi were done on 9-day postnatal

Sprague-Dawley albino rats. The hippocampi were chopped along the transverse axis to 350 µm

thick slices using a McIlwain tissue chopper (The Mickle Latoratory Engineering, Surrey,

England). The slices were placed on 0.4 µm PTFE insert membranes (Millipore, Bedford, MA)

and incubated over 1.2 mL of medium at 36.5 °C in 5 % CO2/ 95 % air for 6 to 8 days. Culture

medium was exchanged every 2 to 3 days. Prior to experiments, culture medium was replaced

with 37 °C GBSS and incubated for thirty minutes. A second GBSS exchange followed for

another incubation period of 30 minutes. A last exchange to 37 °C HBSS and incubation for 30

minutes was done prior to experimental use.

77 5.2.10 Electrokinetic Ejections into Hydrogels

Borosilicate pipettes with filaments (1 mm × 0.58 mm, 4 inches long) (A-M Systems, Inc.,

Carlsborg, WA) were pulled using the Sutter P-2000 capillary puller to tips of approximately 2

µm inner radii. Tip size and shape were measured using an Olympus BX41 optical microscope.

Figure 5.1 shows a 1/4” thick PVC cell with a 1 cm by 1 cm space in its center and a 1 mm wide

× 1 mm high × 2.2 cm long channel milled at its base. The cell was slightly larger than the standard cover slip that was super-glued to its base. The cell was secured on an Olympus IX81 inverted fluorescence microscope stage and imaged with a high resolution charged-coupled device camera (ORCA-ER).

Figure 5.1. Hydrogel cell

A coverslip was super-glued to the base of the cell. An approximately 6 mm thick hydrogel square was placed

in the center space of the cell. A pulled borosilicate pipette filled with fluorophore solution was inserted into

the cell and hydrogel through the channel. On the other end of the channel, a Ag ground electrode was

inserted into the hydrogel.

Hydrogel sections were cut to size and carefully placed in the given space. Careful

attention was made to ensure complete hydrogel contact with the exposed cover slip. The pulled

borosilicate pipette was carefully threaded through one side of the channel and into the hydrogel.

A silver wire ground electrode (0.3 mm diameter) was threaded through the other side of the cell

and was inserted a few millimeters into the hydrogel. The distance between the pipette tip and

the ground electrode was always more than 3.5 mm. A second Ag wire, which is called the

78 working electrode, was inserted into the back end of the pipette. The current source was either a

Digital MidgardTM Precision Current Source iontophoretic pump (Stoelting Co., Wood Dale, IL) or a Princeton Applied Research 173 potentiostat (PAR) (Princeton, NJ). They were attached to the secured electrodes in a circuit containing a relay switch as shown in SI Figure 5.4. Image sequences were acquired using MetaMorph 7.6.2.0 software (MDS Analytical Technologies,

Sunnyvale, CA) using one of the following Olympus objective lenses: UPlan Fl 4×, UPlan S Apo

10×, and a long working distance LUCPlan Fl N 40×. MetaMorph software triggered the onset and cessation of current during image acquisition. The PAR was used in TR70 ejections with currents of 0.1, 0.2, 0.4, 0.5, and 0.6 µA. The PAR setup included a 0.3 mm diameter platinum

(Pt) electrode as the working electrode in order to avoid pipette clogging due to the formation of

AgCl in the pipette. The potential across the system was monitored by either the MidgardTM pump or the PAR potentiostat. Currents ran for 270 to 840 seconds, depending on estimated times to steady-state.

Image sequences were analyzed by drawing a line scan from approximately the center of the pipette tip to the end of the field of view on the same axis as the pipette lumen. Intensity data were recorded along the line scan before the initiation of current (background intensity), and at

30 seconds intervals after the current was applied. Background intensity was subtracted from line scan intensities. Since concentration is related to intensity measured, we can replace concentration, C, in Equation 12 with intensity, I. Reported intensity values are between 0 and

4095. Values of I0, I1, and PeE were determined by nonlinear fitting of Equation 13 to data with

Mathcad 14 (PTC, Needham, MA). I1 is included to correct for baseline intensity.

I( ) = I PeE + I (13) − �휌 1 1−e −Pe 휌 0 � 1−e E �

79 5.2.11 Electrokinetic Ejections into OHSCs

Experiments in the OHSC were set up as shown in Figure 5.2 with the Digital MidgardTM

iontophoresis pump. An insert membrane with an OHSC was placed in a dish with HBSS. A Ag

ground electrode was remotely secured, such that it made contact with the HBSS bath. A pulled

borosilicate pipette filled with fluorophore solution was positioned at a 20° angle using a manipulator arm on a Narishige NMN-21 micromanipulator (Tokyo, Japan). The tip was inserted into the CA1 region of the OHSC. A Ag working electrode was threaded into the distal end of the pipette to make fluid. The electrode circuitry is also the same as the hydrogel experiment. A current of 0.5 µA was applied using the iontophoresis pump for 270 seconds for OHSC experiments.

Figure 5.2. OHSC experimental setup

An OHSC and its attached membrane are placed over HBSS in a dish. A ground electrode is placed remotely in the bath. A working electrode is placed distally in a borosilicate pipette that is inserted into the OHSC at a

20° angle.

OHSCs analyzed for cell death following ejection were subjected to identical conditions,

with the exception that the ejection solution only contained HBSS, without fluorophore, to avoid

interference with propidium iodide (PI) fluorescence. OHSCs were incubated for 24 hours over

culture medium supplemented with PI (final concentration of 7 µM) following current ejection

experiments. Positive cell death controls were subjected to a drop of methanol placed on top of

80 the OHSCs prior to the 24 hour incubation with PI.97 Negative controls were subjected to the

same procedure as experimental OHSCs, without placement of the pipette and application of

current. Following incubation, the medium was exchanged with the same GBSS-GBSS-HBSS procedure described earlier and imaged on the IX81 inverted fluorescence microscope with the

4× UPlan Fl objective lens. PI was imaged using a 560/25 excitation filter, a 625/26 emission filter, both obtained from Olympus, and the Semrock triple-band dichroic mirror described

earlier. A circular region of interest (50 µm2) was used to measure the average intensity within

the area of injection. Cell death was analyzed as described by Hamsher et. al.97 Briefly, PI intensity of OHSCs used in ejection experiments were compared to positive and negative cell death controls. Control OHSC PI intensities were averaged for each experimental day, and compared using Equation 14 to experimental PI intensities of experiments run the same day.

PI intensity (experimental control)- PI intensity (negative control) % Cell Death= ×100% PI intensity (positive control)- PI intenstiy (negative control) (14)

5.2.12 Two-pipette experiments

A second channel was milled perpendicular to the long axis of the cell shown in Figure 5.1, from

the edge of the cell to the center of the square hydrogel pocket. A hydrogel was placed in the

pocket and a pipette filled with TR70 solution was placed as described for the hydrogel

experiments. A counter pipette, filled with G-f HBSS, was placed in the perpendicular channel and inserted into the hydrogel. The pipettes were aligned such that the tips were on the same z- axis plane and 13 µm apart. The control experiment was set up with the ground electrode placed remotely in the hydrogel as described in the hydrogel experiments. The second setup placed the

81 ground electrode in the distal end of the counter pipette, in contact with the solution. 0.5 µA of current was applied for 30 seconds.

5.3 RESULTS

5.3.1 Properties of the fluorophores, hydrogels, and iontophoresis capillaries

Table 5.2 summarizes the diffusion coefficients and mobilities of the fluorophores.

Table 5.2. Fluorophore propertiesa

Observed Free Diffusion Electrophoretic Mobilities Coefficient (D) -9 2 -10 2 (µ ) /(10 m /Vs) (10 m /s) ep TR70 0.37 ± 0.02 0.46 ± 0.03 (N = 32)b (N = 3)b F70 0.38 ± 0.01 -8.80 ± 0.2 (N = 48)b (N = 3)b TR3 1.05 ± 0.05 2.56 ± 0.04 (N = 48)b (N = 3)b 2+ Ru(bpy)3 3.70 ± 0.34 25.86 ± 0.02 (N = 48)b (N = 3)b aEach entry is a mean ± standard error of the mean (SEM). bNumber of replicates.

Electrokinetic flow in the hydrogel is partially governed by the properties of the hydrogel. Table 5.3 shows the ζ-potential and electrokinetic tortuosity of each hydrogel. There is a roughly linear relationship between the w/w percentage of acrylic acid and the ζ-potential. The

25% acrylate-content hydrogel has a ζ-potential similar to the OHSC. Hydrogel tortuosity was more difficult to control than ζ-potential, although consistent within each hydrogel. Tortuosities reported in Table 5.3 are determined by 70 kDa dextran conjugates using an electrokinetic method that measures ζ-potential and tortuosity simultaneously.77, 78 Hydrogel porosity was determined by a series of dehydration experiments. The 25% acrylate-content hydrogel

82 composition had a larger percentage of water than the hydrogels with lower amounts of acrylic

acid. The 25% acrylate-content hydrogel contained 98.3 ± 0.1 % water, while the 10 % and 0 %

acrylate-content hydrogels contained 97.5 ± 0.6 % and 96.6 ± 0.2 % respectively (N = 3 for each

measurement, mean ± SEM). Due to the high porosity values, conductivity was assumed to be

equal to the conductivity of the buffer, which was measured to be 1.62 S/m.

Table 5.3. Hydrogel Properties

% Acrylic Acid ζ-potential 70 kDa Electrokinetic Number of Runs Tortuosity (N) 0 -0.25 ± 0.04 1.69 ± 0.16 4 10 -9.5 ± 0.2 1.61 ± 0.08 4 10 -15.2 ± 0.8 1.46 ± 0.07 5 25* -23.0 ± 0.3 2.36 ± 0.18 4 25 -24.7 ± 1.7 1.71 ± 0.18 9 25 -22.1 ± 0.6 1.07 ± 0.03 6 *Note: the pH for this 25% hydrogel was not adjusted to pH 7-8 before radical polymerization.

Herr et al. were able to quantify iontophoretic solute ejections into free solution relative

to a neutral marker.198 They included a discussion of the effects of ionic strength and pH on the pipette ζ-potential. By accounting for these two parameters, we are able to apply their findings to our electrolyte conditions (G-f HBSS: pH 7.4, ionic strength ≈ 0.15 M). The resulting pipette ζ-

potential of approximately -10 mV agrees with the literature on other forms of borosilicate

glass.203-205 This agrees with our experimental observation that the pipette ζ-potential must be

-9 2 more positive than -11 mV because experiments with F70 (µep = -8.8 × 10 m /Vs) solution do not result in ejections of F70 from the pipette (data not shown).

83 5.3.2 Two-pipette experiment

1000

800

600

400 Intensity

200

0 0 100 200 300 400 Distance (µm)

Figure 5.3. Two-pipette experiment

Figure 5.3 shows the results from the two-pipette experiment in the -23.0 mV ζ-potential hydrogel with 0.5 µA

applied current. In both panels, two pipettes sit perpendicularly to one another. The horizontal ejection

pipette is filled with TR70 solution, and the vertical counter pipette is filled with G-f HBSS electrolyte. The left panel shows an ejection where the ground electrode is placed remotely in the hydrogel. The right panel shows the identical setup with the ground electrode placed in the distal end of the counter pipette. Both images are acquired after 25 seconds of applied current. The scale bar represents 20 µm. Fluorescence

intensity is measured along the dotted lines at the top of both images and is shown in the bottom panel. The

pink line shows the fluorescence intensity corresponding to the left panel, and the purple line corresponds to

84 the right panel. The bottom panel demonstrates that fluorescent material moves into the pipette in a directed

movement with the current flow. Furthermore, there is a greater intensity outside of the pipette in the right panel than in the left.

We tested the idea that a neutral fluorophore can be directed from one pipette to another. Two

iontophoresis pipettes were placed with their tips approximately 13 µm apart in a large ζ-

potential (-23.0 mV) hydrogel (see Figure 5.3). TR70 in the pipette containing the positive working electrode is ejected into the hydrogel. After 25 seconds, the pipette containing the ground electrode shows more fluorescence than the surrounding hydrogel, indicating transport of the fluorophore into the pipette with the ground electrode. However, when the ground electrode is placed in the hydrogel, the same pipette but without the ground electrode has a fluorescence intensity similar to the adjacent hydrogel.

5.3.3 Time to steady-state

Steady-state is defined here as a constant intensity at ρ = 1 for two 30 second intervals. SI Figure

5.1 shows intensity versus distance plots as well as a plot of the intensity versus time of two

experiments. The first experiment reaches steady-state, while the second is approaching steady- state within the same amount of time. SI Table 5.1 gives times to steady-state. The average times to steady-state are 207 and 423 seconds for TR70 and TR3, respectively in the -0.25 mV ζ- potential hydrogel. The time to steady-state correlates with increasing current for TR70 in the -

24.7 mV hydrogel. Some iontophoretic ejections were truncated prior to steady-state to avoid concerns about electrochemical reactions in the pipette and evaporation of water from the hydrogel.

85 5.3.4 Correlation between ζ-potential and solute penetration distance:

Figure 5.4 represents intensity profiles from steady-state iontophoretic ejections of TR70 and

TR3 in hydrogels. Data points are the averages of the dimensionless distance, ρ, from the tip at

which the intensity relative to the intensity at ρ = 1 is 75, 50, and 25 %. The solid lines represent

the intensity profiles (Equation 13, I1 = 0) that pass through the data point for intensity equal to

half of the maximum intensity. The dotted lines denote the intensity profile expected for solute

transport entirely by diffusion from a hemispherical source (C /C0= 1/ρ). Curve fitting of the data

from each experiment to Equation 13 produced correlation coefficients greater than 0.98. Figure

5.4 shows that at the same point in time, the penetration distance of each solute is higher in the

more negative ζ-potential hydrogel. The nearly neutral solute, TR70, shows the degree of EOF

that occurs in each hydrogel (Figure 5.4a). More negative ζ-potentials result in faster EOF and thus a greater penetration distance. In comparison, Figure 5.4b shows this effect on TR3, which is noticeably carried further.

86 100%

80%

60%

40%

% of Maximum Intensity % of Maximum 20% 1 6 13

0% 0 10 20 30 40 50 60 70

Distance, ρ

Figure 5.4a

100%

80%

60%

40%

% of Maximum Intensity % of Maximum 20% 10 12 16

0% 0 50 100 150 200

Distance, ρ

Figure 5.4b

Figure 5.4. Solute penetration distances

Figure 5.4a: Solid lines represent average intensity profiles (t = 270 seconds) from 1 µA ejections of TR70 in hydrogels of varying ζ-potentials: diamond, -0.25 mV (N = 9 intensity profiles); square, -9.5 mV (N = 11 intensity profiles); and triangle, -23.0 mV (N = 9 intensity profiles). Results for the lower ζ-potential hydrogels are at steady-state, while those for the larger ζ-potential hydrogel are approaching steady-state

(row 13 in SI Table 5.1). b: average TR3 intensity profiles (t = 540 seconds) in the following hydrogels:

87 diamond, -0.25 mV (N = 11 intensity profiles); square, -15.2 mV (N = 8 intensity profiles); and triangle, -24.7 mV (N = 6 intensity profiles). The dotted lines show the steady-state diffusion profiles. Labels indicate the corresponding specific row in SI Table 5.1 of the experimental parameters.

5.3.5 Effect of current on concentration and Pe

Figure 5.5 shows the intensity at ρ = 1 and PeE as functions of current with tip size, exposure time, and hydrogel constant. Each data point represents a mean ± SEM of N runs (see SI Table

5.1 for values). An increase in current yields an approximate linear increase in the intensity at ρ

= 1, and thus the concentration of ejected solute.

1000 9 50 7 800 5 40

600 30 E 0 I Pe 400 3 20 2

200 10

0 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

i (µA)

Figure 5.5. Effect of current on concentration and Pe

Figure 5.5: Impact of current on TR70 steady-state values of I0 (represented by diamonds) and Pe

(represented by circles). Ejections were done in the -24.7 mV ζ-potential hydrogel, with the same pipette size and imaged using the same exposure time. Values are taken at time point 270 seconds.

Furthermore, there is a notable approximately linear increase of PeE as a function of current.

Increasing the current increases the penetration of the solute into the hydrogel.

88 5.3.6 Comparison of PeE with PeT

50 12

11 40 9 7

E OHSC 30 8 Pe 5 20 10 3 2 6 10 4 1 0 0 2 4 6 8 10 12 14 16 18 PeT

Figure 5.6. Steady-state PeE vs. PeT

Blue diamonds represent TR70; green squares represent TR3; and the red dot represents TR70 in the OHSC.

PeE was determined by fitting Equation 13 to the experimental steady-state intensities. Equation

5 was used to determine PeT. Figure 5.6 shows that PeE correlates linearly with PeT. The

resulting linear regressions for TR70 and TR3 had respective slopes of 1.7 and 5.9 and

intercepts of 4.2 and 11.4. Labels refer to the row in the supplemental table. Of 120 experiments,

nine outliers were eliminated due to low correlation coefficients from the fit of Equation 13 to

the experimental intensity profiles.

Figure 5.7. Injections into an OHSC

Fluorescence images overlaid on bright field images of the ejection of TR70 into the CA1 region of an OHSC under 0.5 µA. Left: immediately prior to applying current (t = 0 seconds). Right: current has been applied for t = 120 seconds. The scale bar represents 50 µm.

Figure 5.7 shows fluorescence and bright field image overlays of a 0.5 µA ejection of TR70 into the CA1 region of an OHSC at t = 0 and 120 seconds. The radius of the ejected TR70 solution, measured at t = 120 seconds, was ~85 µm. Bright field imaging during the ejection showed that applying 0.5 µA often creates a slight, temporary distortion in morphology of tissue. The tissue appears to recover within 24 hours of the ejection experiment. Ejections into the CA1 regions of

OHSCs (0.5 µA for 300 seconds) showed no increase in cell death by PI staining in comparison to controls (-6 ± 4 % cell death; negative control = 0%, positive control = 100%).

5.3.7 Small, highly charged solute:

2+ We used tris(2,2’-bipyridine)ruthenium, Ru(bpy)3 , to demonstrate the effect of matrix ζ- potential and ionic strength of a solution on a small, highly charged solute. Transport of

2+ Ru(bpy)3 is affected by the hydrogel ζ-potential similarly to the dextran solutes. When

90 2+ distances are compared at 25 % of the maximum intensities, ejections of Ru(bpy)3 solution into

-22.1 mV ζ-potential hydrogel (PeE = 51) reach 1.5 times further than ejections into -0.25 mV ζ-

2+ potential hydrogel (PeE = 24). Ru(bpy)3 was further used to investigate the effect of ionic

strength on the solute ejection. Steady-state ejections within a -22.1 mV ζ-potential hydrogel of

2+ 1.2 mM Ru(bpy)3 in a 5 mM HEPES, 5 mM NaCl buffer (PeE = 88) travels 1.8 times further

2+ than Ru(bpy)3 in a 5 mM HEPES, 150 mM NaCl buffer (PeE = 51).

5.4 DISCUSSION

5.4.1 Fluorophore, hydrogel, and pipette parameters

TR70 and TR3 are commercial products, and as such, their properties are not under our control.

While TR70 is virtually neutral, TR3 is not (refer to Table 5.2). To assess the relative

importance of electrophoresis vs. electroosmosis, we have tabulated the ratio of the ζ-potential of

the solute (related to its mobility as in Equation 4, but without the negative sign) to the ζ- potential of the hydrogels in SI Table 5.1. All of these ratios are significantly greater than unity except for two – TR70 and TR3 in the 0% hydrogel. Thus, with these exceptions, the

observations relating to electric-field-dependent dextran transport are caused by electroosmosis

in the hydrogels. Nonetheless, in discussions below concerning PeT, we do include the

electrophoretic mobility.

We estimate the pipette ζ-potential to be about -10 mV. In situations like this where there

are two media with different ζ-potentials, pressure is created.206 When the hydrogel’s ζ-potential magnitude is small, there is the possibility of transport by pressure-induced flow arising from the

91 larger magnitude of ζ-potential in the pipette. In cases where the hydraulic permeability is low in

the medium with the larger ζ-potential magnitude (pipette), and the hydraulic permeability is

high in the medium with the smaller ζ-potential magnitude (hydrogel), pressure-induced fluid flow in the hydrogel is likely to be small.22 We cannot, however, rule out some contribution from pressure-induced flow.

5.4.2 Time to steady-state

The time to steady-state appears to be roughly correlated with the solute penetration distance for

TR70 for which we have the most data. In an electrochemical context, Zoski et al. have determined the time to reach a given fraction of the steady-state measured quantity, e.g., current or potential, for spherical and disc microelectrodes under diffusion controlled galvanostatic and potentiostatic conditions.207 For a constant current flowing at a hemispherical microelectrode, analogous to the data reported here, the time depends on the closeness to steady-state, the diffusion coefficient of the solute, and the size of the electrode. In the diffusion controlled case, the time to steady-state is proportional to the electrode diameter squared. We can calculate the time to steady-state assuming diffusional mass transport only with the parameters of the experiments discussed here. With a tip (i.e., hemispherical electrode) radius of 2 µm and the diffusion coefficient of TR70, the time to reach 90 % of the steady-state condition is about 3 seconds, and to reach 95 % is about 14 seconds. These times are short in comparison to our observations of transport from a hemispherical source by both diffusion and electrokinetic processes.

92 The length of time to reach steady-state could be exceptionally long with certain

experimental parameters, bringing concern about the effects of the electrochemical processes

occurring at the electrodes and hydrogel/ tissue losing water. To avoid these concerns,

experiments did not exceed 15 minutes. Comsol simulations (see Comsol Simulation section in

SI) indicate the concentration profile at this time is very similar to that of steady-state. Figure 5.4

includes two experimental conditions that were truncated just short of steady-state: TR70 and

TR3 with 1 µA applied current in the largest ζ-potential hydrogels. TR70 and TR3 experiments

are truncated at 270 and 540 seconds.

5.4.3 Solute penetration distance

The ζ-potential of the matrix, magnitude of applied current, pipette properties, and solute

characteristics2 should all affect the solute ejection according to Equation 5. The penetration

distance of the ejection (Figure 5.4) clearly depends on the ζ-potential and thus on the EOF.

Intensity profiles for the 0 % acrylate-content hydrogel (ζ = -0.25 mV) would be expected to be similar to that for diffusion alone in the absence of electrophoresis. However, the penetration distance is larger than that for diffusion for the solute TR70, which has a negligible electrophoretic mobility. It is possible that there is a contribution from pressure-induced flow in this case. For TR3, there is a significant contribution to the mass transport from electrophoresis, as well as a potential contribution from pressure-induced flow.

Unlike hydrogels, most biological matrices cannot be altered with the purpose of varying the magnitude of ζ-potential to increase solute penetration distance. Alternatives to control PeE

are to vary the current applied and the electrolyte concentration. Concentration and PeE are

93 linearly proportional to the current (see Figure 5.5). Thus varying current allows the user to

influence the degree of solute penetration. To our knowledge, many have observed this

phenomenon in brain tissue iontophoresis, but have not shown the quantitative linear correlation

of penetration distance with current.192, 208

The pipette tip opening, tapering, and ζ-potential influence the electric field at the

interface.150, 192 Therefore, it is imperative to measure the pipette opening dimensions prior to

use. The tips are often slightly ellipsoidal, hence we use the average radii of the short (a) and long (b) axes of the pipette tip opening, ā, to represent the radius. Measurements of the tip opening are used in Equation 2 to approximate the electric field at ρ = 1, E0, and thus calculate

PeT. Hence, the hemispherical area applied to PeT at the pipette tip opening is 2πab. Ejection

profiles may also be affected by the pipette ζ-potential. Pipette ζ-potential can be varied with

coatings, or alternatively using another pipette material, such as a fused-silica. Fused-silica glass has a more negative ζ-potential than borosilicate.93, 203, 209

5.4.4 Two-pipette experiments

In a matrix with significant ζ-potential, such as brain tissue or high acrylate-content hydrogel,

EOF can be directed by the current path to desired regions as shown with the two-pipette

experiment. Figure 5.3b shows that the fluorophore is concentrated in the pipette containing the

ground electrode (the ‘counter’ pipette). The concentration of the fluorophore in the counter

pipette cannot occur by diffusion alone. While certainly TR70 can and does diffuse into the

counter pipette, it cannot become more concentrated in the counter pipette (in comparison to the

hydrogel nearby) by diffusion alone. Thus, in the control experiment where the ground electrode

94 is in the hydrogel rather than the pipette there is no concentration of fluorophore in the counter

pipette. Only when current is directed to flow into the counter pipette do we see concentration of

the fluorophore in the counter pipette. As the electrophoretic mobility of TR70 is on the order of

40 times smaller than the electroosmotic mobility in this experiment, the directed movement of

the fluorophore is not by electrophoresis of the nearly neutral TR70 solute. Therefore, the

directed flow is a result of the significant EOF, and thereby ζ-potential of the hydrogel matrix.

5.4.5 PeE versus PeT

Figure 5.6 shows that the measured PeE and the calculated PeT are linearly correlated for the two

dextran conjugates, but the slopes are not unity. Among the factors that may be responsible for

the lack of unit slope are the varying porosities of the media supporting EOF, uncertainty about

the tip size and shape, uncertainly about pressure-induced flow, and the tortuosity of the medium. Porosity is not likely to be a factor. The ratio of PeE and PeT is the same for an OHSC

and hydrogel, indicating the slope is not a result of specifics of the hydrogel, such as porosity. In

simple models of porous media, volume-averaged electrokinetic flow rate is unaffected by changes in porosity.79, 210 This is because the decrease in porosity increases the electric field

(because the current density increases) and thus the local EOF increases within the pores, but at

the same time, the fraction of the porous medium that is fluid, and thus is moving, decreases by

exactly the same factor. Therefore, the OHSC (porosity of ~0.2)18 has Pe values that would also

be expected with the hydrogels (porosity of ~0.98).

The uncertainty about the tip may be a factor in the slope of the plots in Figure 5.6. The

electric field at the tip is approximated a hemisphere in PeT to make the derivation of the steady-

state concentration profile possible. In order to obtain intensity-distance profiles from images, we

95 must accurately find the location of ρ = 1. This is not a simple matter. An error of ± 1 µm in

determining the position of ρ = 1 away from a typical 2 µm radius pipette tip opening provides a

5 to 10% error in PeE. As the location of ρ = 1 is visual, there may be a systematic error the

estimation of that point.

Furthermore, a small pressure effect may arise from a ζ-potential mismatch of the pipette

211 and the hydrogel matrix, which is not accounted for in PeT (refer to SI for further information).

As Table 5.3 indicates, most hydrogels and the OHSC have different ζ-potentials than the

pipette. A pressure effect arises in the case of mismatched ζ-potentials211 because the law of

conservation of mass requires pressure to compensate for differences in EOF. For example, for a

neutral solute being ejected into a neutral gel, PeT is zero. However, the negative ζ-potential in the pipette will create a positive pressure at the tip-hydrogel interface, creating pressure induced flow. The magnitude of the pressure depends on many factors other than the ζ-potentials, so estimating the pressure from parameters of the experiment is difficult. Nonetheless, as there is a pressure-induced velocity when PeT is zero, the y-intercept in Figure 5.6 is likely to be a reflection of the pressure at the pipette/ hydrogel interface and thus the ζ-potential mismatch.

Finally, there is the issue of the tortuosities and related factors of a medium that impede solute motion in comparison to free solution. When referring to tortuosity, we strictly imply an effect of path length. Other factors, such as molecular friction, often characterized by the Ogston model, contribute to the effect of a porous medium on molecular motion.212 To a first

approximation, electrokinetic velocity (the sum of vep and veo) will be impeded by tortuosity to

the same degree as diffusion. Thus, the value of Pe, and therefore the concentration profile,

should not depend at all on tortuosity. It is for this reason that we have not included tortuosity

explicitly in PeT. The slopes of Figure 5.6 give us reason to question this assumption. As is well-

96 known in polyacrylamide gel electrophoresis, the impedance of the gel medium is greater the

larger is an object (at constant ζ-potential of the object) traveling by electrophoresis.212 This is

viewed as a steric exclusion of solute giving rise to the molecular weight dependence of its

mobility. More recently, models and experiments show that the factor by which diffusion in a

hydrogel differs from diffusion in free solution varies strongly with hydrogel porosity in the

range of 0.99 to 0.95 and molecular size especially when the molecular size is greater than the

characteristic dimension of the fibers (i.e., diameter of a polyacrylamide chain) making up the

hydrogel.213 These are the conditions of our experiments. The foregoing tells us that we might expect a molecular-weight dependence and a hydrogel porosity dependence on the apparent tortuosity. In principle, these are questions that are addressable experimentally. However, the

hydrogel porosity is unfortunately correlated with ζ-potential because the more ionic and thus higher ζ-potential hydrogel swells more than the neutral, near zero ζ-potential hydrogel.

Finally, there may be a difference between the electrokinetic and volumetric tortuosities78. Electrokinetic tortuosity is different than volumetric tortuosity due to its

dependence on pore orientation with the electric field79, 80 and hydrogel swelling where the

electric field is high.81, 82 The potentially systematic but unknown errors in determining electric

field (the tip is not a hemisphere), in determining the location of ρ = 1, and pressure effects lead

to a scaling factor that should not depend on molecular weight of the solute. The difference in the

behavior of TR70 and TR3 shown in Figure 5.6 is related to molecular weight, and thus likely

due to the tortuosity effects described above. The role of molecular-weight dependent friction in

the observed phenomena is under investigation.

97 5.4.6 Ejections into OHSCs

Ejections of TR70 solution into the CA1 regions of OHSCs demonstrate significant penetration

distances, (ejection radius of ~85 µm of in 120 seconds as shown in Figure 5.7). The

supplemental table shows the non-steady-state PeE (data points 13 – 16) along with the values for

steady-state (data points 1 – 12). Note that the non-steady-state experimental sets are equally as

reproducible as the steady-state experimental sets, as demonstrated by the PeE SEM displayed in

SI Table 5.1. PeE of TR70 ejections in OHSCs fall directly in line with PeE determined in hydrogels, indicating the hydrogels are viable models for OHSCs. Hydrogels with ζ-potentials

around -22 mV were originally meant to model the OHSC that has a known ζ-potential of -22.8

± 0.8 mV and an electrokinetic tortuosity of 2.24 ± 0.10.77 However, ejections in OHSCs show

an increased time to reach steady-state than in comparable, 25% acrylate-content hydrogels.

Although ejections in OHSCs appear to be approaching steady-state after 270 seconds, the

ejection profile continues to grow slightly for more than an hour. As the OHSC is only about 150

µm thick, when the distance traveled reaches distances on the order of 100 µm (ρ ~50) we can no

longer assume a spherically symmetrical field or concentration profile. This topic must be further

investigated to make any quantitative conclusions.

Propidium iodide analysis demonstrates that 0.5 µA ejections for fewer than 300 seconds

into the CA1 region do not increase cell death in the OHSC. However, there is a slight distortion

of the morphology around the ejected solute that is largely resolved within 24 hours following

ejections. The ejection plume may also be slightly asymmetrical due to any anisotropy and

heterogeneity within the OHSC.

98 Solutes can be transported further by decreasing the electrolyte concentration. Many

researchers use low ionic strength solutions for iontophoretic delivery of small charged

molecules.18, 98, 198 To our knowledge, there is no standard for electrolyte concentration.198 The

electrical double layer thickness is proportional to the reciprocal square root of ionic strength.

Consequently, the absolute value of the ζ-potential increases with decreasing ionic strength.

2+ Hence, the Ru(bpy)3 solution containing 5 mM NaCl creates a larger ζ-potential and

consequently the solute ejection distance is 1.8 times farther than the 150 mM NaCl solution.

5.4.7 Conclusion

The major conclusion is that under appropriate conditions, the transport of a solute into a porous,

aqueous medium depends strongly on the ζ-potential of the medium. We demonstrated that solute ejection distance is affected by the solute, solute-containing medium ionic strength, and hydrogel properties, as well as the current applied. PeE, which relates to solute penetration

distance, is linearly correlated to PeT, which relates to measureable parameters. OHSCs have no

added cell death following long-term electrokinetic ejection. Thus, this technique is applicable to

living tissue for long-distance solute transport. Furthermore, results of two-probe experiments

indicate that the EOF path can be directed in tissue using placement of the counter electrode.

5.4.8 Acknowledgement

We thank the NIH for support through grant GM44842.

99 5.5 SUPPORTING INFORMATION

500 30 2500 30s 60 60s 400 90 2000 90s 120 120s 300 150 1500 150s 180 180s

200 210 Intensity 1000 210s

Intensity 240 240s 100 500 270 270s

0 0 0 20 40 60 80 0 20 40 60 80 -100 ρ ρ

1400

1200

1000

800 Intensity 600

400

200 0 30 60 90 120 150 180 210 240 270 300 Time (s)

SI Figure 5.1. Steady state versus non-steady state

SI Figure 5.1 shows 30 second interval ejection profiles of TR70 under 1 µA current in hydrogels with ζ-potentials of -9.5 (left panel) and -23.0 mV (right panel). The left panel reaches steady state by t = 270 s of applied current, while the right panel does not. The bottom panel is a plot of the intensity at ρ = 75% of the maximum intensity versus time. The red squares relate to the left panel that reach steady state as demonstrated by a slope of 0 over several time intervals.

100 The blue diamonds are associated with the right panel corresponding data. The right panel

intensities are increasing steadily, however the slope is decreasing over time, indicating that the

experiment is approaching steady state.

SI Table 5.1. Experiment summary table

Solute Data Hydrogel ζhydrogel Current Average Time to PeE SEM N Point ζ-potential (mV) applied (µA) Steady State (s) ζsolute TR70 1 -0.25 � 0.43 � 1 207 6 0.74 9 TR70 2 -24.7 43 0.100a 366 15 0.47 3 TR70 3 -24.7 43 0.200 a 492 18 0.88 5 TR70 4 -9.5 16 0.500 240 12 0.34 5 TR70 5 -24.7 43 0.400 a 780 27 0.76 4 TR70 6 -9.5 16 1 233 13 2.1 11 TR70 7 -24.7 43 0.500 a 953 36 1.5 4 TR70 8 -24.7 43 0.500 818 30 0.95 4 TR70 9 -24.7 43 0.600 a 1030 39 4.6 3 TR3 10 -0.25 0.077 1 423 16 0.36 11 TR3 11 -15.2 4.7 0.500 498 37 1.3 13 TR3 12 -15.2 4.7 1 510 51 1.3 8 Solute Non-steady State Data TR70 13 -23.0 40 1 - 27 1.4 9 TR3 14 -22.1 6.8 1 - 52 3.2 10 TR3 15 -24.7 7.6 0.5 - 69 4.9 6 TR3 16 -24.7 7.6 1 - 73 2.5 6 aCurrent applied through the PAR circuit. All other experiments were run using the Digital MidgardTM

Precision current source.

The ζ-potential of the solute is calculated by SI Equation 1 using the electrophoretic mobility,

µep, and the permittivity (ε) and viscosity (η) of the water.

ζsolute = (S1) η ε ep µ

101 5.5.1 Comsol Simulations:¥

¥Comsol simulations were primarily completed by Manyan Wang in the Weber lab.

A commercial finite element method (FEM) was used to calculate simulations with Comsol 3.5a

(Comsol, Burlington, MA). The simple model consisted of a free flow pipette (height of 100 µm,

radius of 2 µm) adjacent to a large porous hydrogel (height of 5 mm, radius of 5 mm). A two

dimensional axis of symmetry was drawn through the pipette lumen. The steady-state solution

was the result of three descriptive modes: “conductive media DC” mode, “Brinkman equation” mode, and “electrokinetic flow application” mode. The conductive media mode solves Equation

S2 with appropriate boundary conditions

( V J ) = 0 (S2) e w∇here∙ σ ∇V is− potential, σ is conductivity and J is an externally generated current density. An e inward current flow of 50000 A/m2 was set at the distal end of the pipette. The result is the

potential distribution for a given J . e We make the following assumptions in the Brinkman equation mode: a low Reynolds

number, low flow speed, and small pores. This mode solves the continuity Equation S3 and a

momentum balance Equation S4.

= 0 (S3)

∇ ∙ 퐮 + = + ( + ( ) ) + (S4) 휌 휕퐮 휂 휂 푇 휙 휕푡 휅 퐮 ∇ ∙ �−푝퐈 휙 ∇퐮 ∇퐮 � 퐅 where u is the velocity vector, ρ is the density, porosity is , t is time, represents the dynamic

viscosity, p is the pressure, the identity matrix is described휙 by I, (∇u)휂T represents the matrix

transpose, F is the body force vector, and is the permeability of the porous medium calculated from the Equation S5. 휅

102

= (S5) 2 휙 푚 휅 푘0 휏 In Equation S5, m is the pore radius, is a factor of the pore shape, and is the volumetric

0 tortuosity of the medium. Volumetric tortuosity푘 is related to the physical structure휏 of the porous volume. In order to calculate the electroosmotic flow inside the pipette, the pipette was defined as a moving wall, where velocity was calculated by Equation S6.

ζwall wall = (S6) 휀 푣 − 휂 퐸 The electroosmotic flow in the porous gel was set as a body force calculated by Equation S7

ζ = hydrogel (S7) 휀 2 퐅 − 푚 휏 퐸 where ζwall is the ζ-potential of the pipette wall, ζhydrogel is the ζ-potential of the hydrogel and E

is the electric field obtained from conductive media DC mode.

The electrokinetic flow application mode solves the Nernst-Planck equation (S8) with a continuity equation (S9).

C + C + C + C = 0 (S8) 휕 휕푡 푒푝 ( ∇ ∙ �−)퐷=∇ 0 휇 푬 퐮′� (S9)

1 2 Diffusion퐧 ∙ 퐉 − 퐉 coefficient of the solute is D, C is the concentration of the solute, and µep is the

electrophoretic mobility of the solute. The velocity, , is equal to the velocity of the solvent

obtained from the Brinkman equation divided by the퐮 ′square of the apparent tortuosity of the

solute in the gel (obtained from the experiments). The solute mass flux vectors on either side of

the hydrogel/solution interface, J1 and J2, are defined by Equation S10.

= C + C + C (S10)

퐉 −퐷 ∇ 휇푒푝푬 퐮′

103 The solute contribution to the solvent velocity is assumed to be negligible in comparison to the solvent’s contribution to the solute. Note that the boundary condition at the interface of the pipette and the hydrogel was set as a continuity condition in all three modes. The mesh size was set to a maximum element size scaling factor of 0.3 and narrow regions setting of 5, resulting in

172490 degrees of freedom. The mesh size creates a stable solution without too many elements.

The ‘parametric segregated’ stationary solver was chosen for rapid convergence of the calculation.

We also simulated a more realistic geometry using the same three modes as in the simple simulation with a few exceptions. The more realistic simulation calculated the parameters of a pulled pipette tip in -25 mV ζ-potential hydrogel under 1 µA of current. The pipette was drawn using 24 points measured from a micrograph of a typical pipette. The tip opening was rounded in order to avoid problems with the mesh. The simulated pipette consisted of a 0.5 mm non-tapered section and a 2.5 mm tapered section. The hydrogel was represented by a cylinder with a radius of 10 mm and a height of 13 mm. The pipette, filled with free solution, was placed so that the tip was 3 mm below the top surface of the hydrogel. Hydrogel also filled the tip to a depth of 0.5

µm. We applied the default extra fine mesh with the additions of a maximum element size of 5 ×

10-5 in the pipette, a maximum element size of 2 × 10-6 around the tip, and a growth rate of 1.1 around the tip, resulting in 122348 degrees of freedom.

104

SI Figure 5.2. Comsol Simulation

SI Figure 5.2 shows the result of a simple Comsol simulation of a 2 µm radius pipette (ζ

= -25 mV) ejecting a solute into a hydrogel (ζ = -25 mV, λ = 1, pore radius = 200 nm) via an applied current density of 5 × 104 A/m2 at the distal end of the pipette. Arrows show the direction and magnitude of the solute flux in the hydrogel. The solute has no electrophoretic mobility and a diffusion coefficient of 1 × 10-10 m2/s. The simulation is pseudo-3-dimensional simulation of the simplified version of our experiment. The concentration profile along the axis of symmetry yields a concentration of approximately 1 at ρ = 1 and a PeE of 10. PeE is determined by fitting

Equation 13 to the concentration data along the axis of symmetry outside the tip. By applying the above parameters to Equation 5, the resulting PeT is 10. Simplified simulation models such as shown here demonstrate the agreement between PeE and PeT.

105 A scaling factor between the two Pe values occurs when there is a difference in the ζ-

potentials of the pipette and the hydrogel, as shown in SI Figure 5.3.

4 E Pe

0 -1.0 -0.5 0.0 ζ ζ ζ ( pipette - gel)/ gel

SI Figure 5.3. Effects of mismatched ζ-potentials

SI Figure 5.3 shows the simulated effect of a ζ-potential mismatch at the pipette/ hydrogel interface on PeE.

The solid red line refers to a -25 mV hydrogel and the dashed blue line refers to a -10 mV hydrogel. The pipette ζ-potential is varied for each line. Tortuosity is 1 in both hydrogels.

SI Figure 5.3 shows PeE of a neutral solute as a function of the ζ-potential mismatch:

ζ ζ pipette hydrogel (S11) ζpipette−

In the case where the pipette and hydrogel ζ-potentials match, the ζ-potential mismatch is zero.

As the magnitude of the pipette ζ-potential approaches zero, the ζ-potential mismatch approaches -1. The figure shows the relationship between PeE of the solute and the ζ-potential

mismatch is linear. The slope and intercept depend on the ζ-potential of the hydrogel. The larger

slope and intercept result from the larger magnitude ζ-potential (-25 mV) hydrogel. These simplified geometry simulations exhibit increased pressures at the pipette/ hydrogel interface with greater ζ-potential mismatches. In a -25 mV hydrogel, for example, pressures of 32 (-10

106 mV pipette ζ-potential, mismatch= -0.60) and 47 Pa (0 mV pipette ζ-potential, mismatch= -0.96) are produced at the interface. A -10 mV pipette ζ-potential is approximately that of borosilicate.

However, in a more realistic simulation that accounts for tip tapering, the resulting

pressure is 6 Pa. The pressure is proportional to the taper length, defined as the distance between

the unpulled pipette diameter and the opening at the tip. Increasing the taper length ultimately

increases the pressure. Note that typical iontophoretic pipettes have a taper that is at least five

times longer than the ones used in our experiments and that were simulated here.. Thus, typical

iontophoretic experiments would likely have greater pressure-induced effects under the condition

that all other variables remain the same. This difference in tapering length is a result of the

method used to pull the pipette to a tip. The P-2000 capillary puller forms reproducible pipette

tips with the use of a laser and a controlled pull velocity. Pressure effects described in the

simulation are somewhat inaccurate. The simulations do not account for temperature effects

because of the added complexity. The simulations are influenced by temperature.

107

SI Figure 5.4a. Digital MidgardTM experimental circuit

108

SI Figure 5.4b. PAR experimental circuit

109 6.0 A SIMPLE METHOD FOR MEASURING ORGANOTYPIC TISSUE SLICE

CULTURE THICKNESS

ABSTRACT

This paper presents a simple method to measure tissue slice thicknesses using an ohmmeter. An ohmmeter can detect when a circuit is open or closed. The circuit described here is composed of a metal probe, an ohmmeter, a counter electrode, culture medium or physiological buffer, and tissue. The probe and the electrode are on opposite interfaces of an organotypic hippocampal slice culture. The circuit closes when the metal probe makes contact with the top surface of the tissue. The probe position is recorded and compared to its position when it makes contact with the insert membrane on which the tissue grows, thus yielding a thickness measurement. The method does not reduce the viability of tissue cultures. Thicknesses of the tissue cultures were measured under a number of culturing protocols. An initial drop in thickness occurred between 0 and 4 days in culture. Thicknesses are rather constant thereafter. The type of culture medium and the initial thickness of the tissue explant influence the thickness. The measurement technique described is straightforward and rapid, about one minute per culture.

110 6.1 INTRODUCTION

Brain tissue slices and slice cultures are frequently used for neuropharmocological and

neurophysiological studies 214. Acute and cultured tissue slices provide easy access for

experimental manipulation while maintaining in vivo morphology and electrophysiology 152.

Freshly excised tissue is sliced to a desired thickness using either a tissue chopper or a vibratome

and is used for experiments immediately, or cultured and used later. The thickness of the tissue

slice is essential to its characterization. The overall health 18, the approach to making certain

measurements 93, 215-217, and diffusion distances 214, 216, 218-222 are directly correlated to tissue

thickness, and are pertinent to neurological experiments. This paper presents a simple, direct, and

cost-effective method for measuring thicknesses of tissue slice cultures without causing cell death.

Methods exist for measuring slice thickness; however, those methods involve either sacrificing the tissue,84, 223 or submerging the tissue in solution219. In the first case, the tissue is cut perpendicular to the long axis of the tissue and the thickness is measured optically. In the second case, double-barreled ion-selective microelectrodes measure steep changes in ion concentration gradients once in contact with the tissue 219. In the latter procedure, the acutely

prepared hippocampal tissue slice is in a bath. The microelectrode detects ion concentration

gradients which change significantly at the bath-tissue interface due to the difference between the non-tortuous bath and tortuous tissue. Although a useful technique, translating this technique to slices cultured using the interface method would be cumbersome. In the interface environment, oxygen is supplied from the air above the tissue culture and nutrients are supplied from below.84 Submerging tissue cultures may therefore cause hypoxia if the medium is not

111 oxygenated. There is therefore a need for a simple and safe method for determining the thickness

of tissue cultured by the interface method.

This paper describes experiments using organotypic hippocampal slice cultures that are

commonly used for neurological studies.152, 214, 224 The hippocampal laminar architecture is

preserved, while the biochemistry, synaptic connections, and electrophysiology are similar to

those in vivo.214 Organotypic cultures survive from weeks to months, providing options for long term studies, such as those studying recovery following excitotoxicity, ischemia, and traumatic

injury.224 Recently, we have developed a technique to acquire samples of extracellular fluid from

these tissue cultures based on electroosmotic flow.93, 215 In this or any other approach to obtain

information on the concentrations and distribution of freely diffusing solutes, knowledge of the

culture’s thickness is essential.

6.2 MATERIALS AND METHODS

6.2.1 Chemicals and solutions

Two types of culture media were employed: medium A and medium B. Medium A78 consisted

of the following obtained from Sigma-Aldrich: 50% basal medium eagle, 25% Earle’s balanced

salt solution, 23% heat inactivated horse serum, 25 U/mL penicillin-streptomycin, 1 mM L-

glutamine, and 41.6 mM D-(+)-glucose. Medium B 225 was composed of the following materials

obtained from Gibco (Invitrogen, Carlsbad, CA): 50% opti-MEM, 25% horse serum, and 25%

Hanks’ balanced salt solution with phenol red, and 1% D-(+)-glucose (from Sigma-Aldrich, St.

Louis, MO). Medium B was filtered through a Nalgene 0.1 µm polyethersulfone filter (Fisher

112 Scientific, Waltham, MA). The two culture media were refrigerated until use, at which point they

were warmed to 37 °C. The following were also purchased from Sigma-Aldrich unless stated otherwise. Gey’s balanced salt solution (GBSS) was supplemented with D-(+)-glucose (27.5 mM) and MgSO4 (2.7 mM, filtered). GBSS was refrigerated until use. HEPES buffered salt

solution (HBSS) contained (mM) 143.4 NaCl, 5.0 HEPES, 5.4 KCl, 1.2 MgSO4, 1.2 NaH2PO4,

2.0 CaCl2, and 10 D-(+)-glucose diluted with Millipore Synthesis A10 system 18 MΩ purified

water (Millipore, Billerica, MA). The pH was adjusted to 7.40 using 0.5 mM NaOH solution

prior to filtration through a 0.45 µm nylon Whatman filter (Maidstone, England). HBSS was

frozen until use, at which point it was warmed to 37 °C. Propidium iodide (PI) solution was

prepared at a final concentration of 0.35 mM in GBSS and frozen until use. 3% agarose type-V

(Sigma-Aldrich) in Hanks’ balanced salt solution with phenol red was used for method

validation.

6.2.2 Tissue culturing procedure

Organotypic hippocampal slice cultures were prepared as described previously.78 The University

of Pittsburgh IACUC authorized the following procedures. Hippocampi were excised from

seven-day postnatal Sprague-Dawley albino rats. The tissue was cut into slices (at different

thicknesses, as indicated) using a McIlwain tissue chopper (The Mickle Laboratory Engineering,

Surrey, England). Tissue slices were laid on a porous Millicell insert membrane (Millipore,

Billerica, MA) and incubated at 37 °C in 95% air/5% CO2 over either medium A or B for up to

14 days in culture. Medium was exchanged twice per week.

113 6.2.3 Tissue thickness measurements

An insert containing one or two tissue slices was placed over culture medium on an inverted microscope (Olympus IX-71, Melville, NY) in order to monitor visually the procedure under bright field. The inset of Figure 6.1 shows the measurement setup. A 1 mm diameter platinum wire was submerged in the culture medium and connected to a multimeter (Fluka, Everett, WA) to monitor resistance. The other terminal of the multimeter was connected to a 1 mm diameter stainless steel rod. One end of the rod was polished to create a flat surface. The rod was lowered using an electronic micromanipulator (MP-285, with ROE-200 controller and MPC-200,

Sutter, Novato, CA) until its polished end made contact with either the insert membrane or the tissue. Contact was indicated by the multimeter reading changing from “open circuit”, to less than 16 MΩ, indicating a completed circuit. Contact and rod location were visually confirmed using the microscope. At the point of contact, the resistance and manipulator position were recorded. Reproducibility of the micromanipulator was tested by recording the position after making contact with the fluid bath, and repeating eight times. For all tissue measurements, the probe was centered primarily on cell bodies of the trisynaptic loop or on apical dendrites of the pyramidal cells, near the pyramidal cell layer. Each tissue culture thickness was calculated from the difference in the averages of four readings from the surface of the insert membrane readings and four tissue readings. With one exception, thicknesses of ten or eleven tissue cultures were measured and averaged for each day in culture. Twenty-five cultures were measured for day zero in medium A. All day zero tissue culture thicknesses were measured 30 minutes to 2 hours following explantation. Errors reported in Figure 6.1 are the standard error

114 of the mean, SEM, for each day in culture. Standard deviations reported in Table 5.1 are the

standard deviations of the measured values at each condition.

To compare the proposed method to a known method of measuring thickness, 5 (400

µm initial slice thickness) and 3 (350 µm initial slice thickness) were cultured for 5 days. On day

5 in culture, each of the tissue culture thicknesses was measured using the probe. The 8 tissues were divided into 3 groups. The first group of tissue cultures (N = 3 tissue cultures) was separated from any excess membrane using an X-Acto knife, and the tissue cultures were sliced to 600 µm thick along the same axis as CA1 axons using the tissue chopper. Each tissue culture yielded approximately 4 slices that contained at least one of the following regions: dentate gyrus, CA1-CA4, or the subiculum. The slices were set on a coverslip along the freshly sliced edge and imaged over the IX-71 microscope under bright field using an Olympus U Plan Apo 4×

objective lens. The second group of tissue cultures (N = 3 tissue cultures) were fixed using 4 %

paraformaldehyde in PBS (Boston Bioproducts, Ashland, MA) for 1 hour, after which the tissue

cultures were sliced and imaged as in the first group. Tissue cultures in the third group (N = 2

tissue cultures) were sliced, then fixed in paraformaldehyde and imaged. Each image yielded 3

to 8 thickness measurement and measurements of the insert membrane thickness. The

thickness measurements of all the sections of the same tissue culture were averaged and

compared to the measured thickness of the same tissue culture using the probe method.

6.2.4 Viability measurements

Thirty tissue cultures were selected at random to check if the process of measuring tissue

thickness compromised viability. Following the thickness measurements, medium was

115 exchanged with fresh 37 °C medium supplemented with PI solution, to a final PI concentration of 7 µM. Following a 24 hour incubation period, the medium was exchanged with 37 °C GBSS

and incubated for 30 minutes. The GBSS was exchanged and incubated again for a second 30

minute interval. A final exchange to 37 °C HBSS and incubation for 30 minutes preceded

imaging analysis. Negative control tissues followed the same exchange procedure. The negative

control tissues remained in the incubator following exchanges. To serve as a set of positive

controls, a drop of methanol was placed on top of tissue cultures prior to PI exposure. Methanol

ensures cell death of both glia and neurons throughout the tissue culture.215 PI fluorescence

indicating cell death was imaged using a U-MGIW2 cube (Olympus) and Simple PCI software

(Compix, Cranberry, PA). Exposure times were set according to the fluorescence intensity of the

positive controls. Analysis of tissue culture images were done with Simple PCI. Average

intensities of regions of interest encompassing the laminar structure of the tissue cultures were

collected and analyzed using Equation 1.215 The intensity results for all the controls were

averaged within their given treatments, resulting in an average PI intensity for the specific day in

culture.

PI intensity (experimental )-Average PI intensity (negative control) % Cell Death= ×100% Average PI intensity (positive control)- Average PI intenstiy (negative control) (1)

6.2.5 Method validation

Agarose gel (3% w/v) was cut to approximately 3.5 mm thick slab. The gel was laid on its end on a coverslip over the microscope and was imaged in bright field. Several thickness

measurements were made at different points along the length of the slab. The procedure was

repeated for the same gel slab on the opposite face. The gel was then laid on an insert membrane

116 over HBSS, like the tissue setup. The thickness of the slab was measured by averaging manipulator readings from 4 points of contact with the steel rod, and subtracting from the readings of 4 points of contact with the membrane. Once complete, two sides of the slab were again imaged and measured using Simple PCI software on a coverslip.

6.3 RESULTS

6.3.1 Tissue culture thickness

Figure 6.1. Measured thicknesses over time

Points show the average thicknesses on the specified days in culture and the standard errors of the mean.

Purple exes represent medium A tissues with an initial slice thickness of 500 µm. Red squares represent medium B and 300 µm initial slice thickness. The blue diamonds represent medium B and 400 µm. The green triangles represent medium B and 500 µm. Inset: The setup of the measurement circuit.

117 Table 6.1. Values corresponding to Figure 6.1

Medium A

Slice Thickness (µm) Day Average Thickness (µm) SEM (µm) n Standard Deviation (µm)

500 0 398 11 25 56 500 1 298 14 10 45 500 2 233 20 10 63 500 3 156 10 10 32 500 4 158 7 10 22 500 5 140 11 10 35 500 6 148 15 10 47 500 7 138 13 10 40 500 8 158 13 10 41 500 9 166 14 10 44 500 10 152 15 10 46 500 11 171 7 10 22 500 12 161 9 10 27 500 13 165 8 10 24 500 14 181 8 10 25 Medium B 300 4 135 7 10 22 300 6 137 9 10 30 300 7 186 9 10 27 300 8 178 8 10 25 300 10 139 16 10 52 300 12 159 7 10 23 300 14 151 10 10 30 400 4 165 9 10 30 400 6 194 8 11 28 400 7 229 9 10 29 400 8 186 8 10 24 400 10 154 11 10 34 400 12 177 9 10 28 400 14 169 7 10 22 500 0 301 25 10 78 500 2 176 10 10 31 500 4 184 6 10 20 500 6 169 9 10 27 500 7 207 8 11 25 500 8 204 10 10 31 500 10 189 6 10 18 500 12 201 4 10 13 500 14 189 7 11 23

118

The proposed method measured tissue culture thicknesses to be 40 ± 9 µm (N = 8) more thick

than when the same tissue cultures were sacrificed and measured in bright field images. Figure

6.1 shows the average thicknesses of tissue cultures for three different initial slice thicknesses

(set by the tissue chopper) and two different culture media. See Table 6.1 for a table list of values

corresponding to Figure 6.1. Day 0 tissue culture thicknesses were measured 30 minutes to 2

hours following explantation. Tissues cultured in medium A were measured every day in culture

from day 0, directly after tissue extraction, until day 14. The other points represent tissues

cultured in medium B with different initial slice thicknesses. Tissue culture thickness levels off

between days 4 and 14, during which time the average tissue thickness is 158 ± 3 µm for

medium A with an initial slice thickness of 500 µm. Medium B produces average tissue

thicknesses for days 4 – 14 in culture of 155 ± 4, 182 ± 4, and 192 ± 3 µm, with initial slice thicknesses of 300, 400, and 500 µm, respectively. The two different culture media result in statistically different tissue culture thicknesses over this time period, with p < 0.0001. Single- factor ANOVA of 500 µm initial slice thickness cultured over either medium shows that there is a slight dependence of thickness on the day in culture. Single-factor ANOVA shows that initial slice thickness affects medium B tissue culture thickness (p < 3 × 10-10).

On a given day in culture, thicknesses can vary as evidenced by the standard deviations in

Table 6.1. Thicknesses of tissues cultured in medium B are less variable than thicknesses of

slices cultured in medium A. For example, for a slice thickness 500 µm on day 6 in culture (n =

10), the SEM in tissue thickness for medium B, 5 % of the mean, is smaller than those cultured

in medium A, 10 % of the mean. Positioning of the micromanipulator includes a 14 µm standard

deviation (n = 8).

119 6.3.2 Viabilty

An average of 3.9 ± 1.2 % cell death is incurred in the tissue cultures as a result of thickness

measurements as compared to negative and positive controls.

6.3.3 Method validation

Images of the gel slab taken before probe thickness measurements resulted in a thickness of 3.44

± 0.03 mm. The slab thickness measured using the proposed probe technique was 3.51 ± 0.05

mm. The thickness measurement of the slab using image analysis following the probe

measurements was 3.52 ± 0.02 mm.

6.4 DISCUSSION

Current methods available for measuring OHSC thickness have significant drawbacks. Most

OHSC users chop the OHSC perpendicularly to the insert membrane in order to image and optically measure the thickness. When we made an attempt using this method in our laboratory, we found it to be unreliable. The tissue cultures were consistently measured thinner once chopped compared to when measured with the probe. The primary concern relates to “pinching” of the OHSC along the chopped edges. Furthermore, manipulation of the chopped OHSCs is challenging. The chopped OHSC has a tendency to dry out or not lay perfectly perpendicular to the coverslip. As stated earlier, the use of ion-selective microelectrodes is cumbersome, and therefore was not used as a method of validation. The use of a 3% gel slab allows for optical

120 measurements without risking dehydration of the gel. In order to position the slab on its end for optical measurements, the gel thickness must be large enough to stand on its side. Gel slabs of similar thickness to OHSCs tend to fall over during optical measurements. In addition, the gel composition consists of Hanks’ balanced salt solution, such that the multimeter responds when the stainless steel probe contacts the surface of the gel. Optical thickness measurements confirmed thickness measurements made with the stainless steel probe, thus validating the proposed method of thickness measurements. Some error in the measurements is due to the 14

µm standard deviation of the manipulator positioning.

Figure 6.1 shows a significant loss of thickness within the first few days. Measurements taken within a couple of hours following explantation show that the tissue thickness is 20%

(medium B) to 30% (medium A) less than the chopper setting. There is no observable dependence of tissue culture thickness on time within the 30 minutes to 2 hour time-span following explantation. These results show thinning occurring on a shorter timescale compared to Stoppini’s original estimation of one week..84 The two media produce slightly different tissue thicknesses once the tissue thicknesses level off. The different thicknesses may arise from differences in glial cell proliferation.158, 226 Visual inspection confirms a small extension of glia that line the edges of the tissue cultures84 in medium B compared to medium A.

The standard deviations in Table 6.1 show variation in thicknesses among tissue cultures.

Variations among the tissue cultures could be due to varying cell differentiation rates and glial cell proliferation,152 since there is a dependence on the medium. Once in culture, reactive astrocytic glia blanket the surface of the sectioned tissue.224, 226 Other factors affecting tissue thickness are related to surgery and incubation. Some of those factors are variations in time spent outside of the culture environment and degree of oxygenation.

121 Measuring tissue thickness is effective for screening and characterizing brain tissue

cultures prior to use in experiments. Brain tissue slices or cultures that are of thicknesses other

than expected are likely unhealthy tissues.18 The technique can be useful in real time measurements of tissue swelling under various stress conditions, as opposed to final characterization methods requiring sacrifice of the tissue culture.84, 223 Characterizing tissue cultures prior to experiments also allows for proper placement of sampling and measurement probes and capillaries.93, 215 Furthermore, sampling analytes from the extracellular space is

affected by tissue thickness.93, 215, 216, 221 Variations in thicknesses among the tissue cultures

strengthen the argument that tissue culture thickness should be measured.

The method proposed in this paper is easy, effective, expeditious, and inexpensive. It

causes negligible cell death, and does not require the tissue cultures to be submerged in a bath.

Furthermore, none of the components are fragile. This simple technique may be applied for

characterization and screening of any tissue slice or culture that is not submerged in solution.

122 REFERENCES

1. Savtchenko LP, Kulahin N, Korogod SM, Rusakov DA. Electric fields of synaptic

currents could influence diffusion of charged neurotransmitter molecules. Synapse.

2004;51(4):270-278.

2. Anastassiou CA, Montgomery SM, Barahona M, Buzsaki G, Koch C. The effect of

spatially inhomogeneous extracellular electric fields on neurons. Journal of

Neuroscience. 2010;30(5):1925-1936.

3. Jefferys JG. Nonsynaptic modulation of neuronal activity in the brain: electric currents

and extracellular ions. Physiological Reviews. 1995;75(4):689-723.

4. Perlmutter JS, Mink JW. Deep brain stimulation. Annual Review of Neuroscience.

2006;29:229-257.

5. Kim EJ, Kim WR, Chi SE, et al. Repetitive transcranial magnetic stimulation protects

hippocampal plasticity in an animal model of depression. Neuroscience Letters.

2006;405(1-2):79-83. Epub 2006 Jul 2012.

6. Kuroda Y, Motohashi N, Ito H, et al. Effects of repetitive transcranial magnetic

stimulation on [11C]raclopride binding and cognitive function in patients with

depression. Journal of Affective Disorders. 2006;95(1-3):35-42. Epub 2006 Jun 2015.

123 7. Lisanby SH, Belmaker RH. Animal models of the mechanisms of action of repetitive

transcranial magnetic stimulation (RTMS): comparisons with electroconvulsive shock

(ECS). Depression and Anxiety. 2000;12(3):178-187.

8. Wagner T, Gangitano M, Romero R, et al. Intracranial measurement of current densities

induced by transcranial magnetic stimulation in the human brain. Neuroscience Letters.

2004;354(2):91-94.

9. Ozen S, Sirota A, Belluscio MA, et al. Transcranial electric stimulation entrains cortical

neuronal populations in rats. Journal of Neuroscience. 2010;30(34):11476-11485.

10. Ardolino G, Bossi B, Barbieri S, Priori A. Non-synaptic mechanisms underlie the after-

effects of cathodal transcutaneous direct current stimulation of the human brain. Journal

of Physiology. 2005;568(Pt 2):653-663. Epub 2005 Jul 2021.

11. Fregni F, Boggio PS, Nitsche MA, Rigonatti SP, Pascual-Leone A. Cognitive effects of

repeated sessions of transcranial direct current stimulation in patients with depression.

Depression and Anxiety. 2006;23(8):482-484.

12. Juckel G, Mendlin A, Jacobs BL. Electrical stimulation of rat medial prefrontal cortex

enhances forebrain serotonin output: implications for electroconvulsive therapy and

transcranial magnetic stimulation in depression. Neuropsychopharmacology.

1999;21(3):391-398.

13. Weinstein S. The anticonvulsant effect of electrical fields. Current Neurology and

Neuroscience Reports. 2001;1(2):155-161.

14. Garcia PA, Rossmeisl JH, Jr., Neal RE, 2nd, et al. Intracranial nonthermal irreversible

electroporation: in vivo analysis. The Journal of Membrane Biology. 2010;236(1):127-

136.

124 15. Akaneya Y, Jiang B, Tsumoto T. RNAi-induced gene silencing by local electroporation

in targeting brain region. Journal of Neurophysiology. 2005;93(1):594-602.

16. McQuiston AR. Cholinergic modulation of excitatory synaptic input integration in

hippocampal CA1. The Journal of Physiology. 2010;588(Pt 19):3727-3742.

17. McCaig CD, Sangster L, Stewart R. Neurotrophins enhance electric field-directed growth

cone guidance and directed nerve branching. Dev. Dyn. 2000;217(3):299-308.

18. Sykova E, Nicholson C. Diffusion in brain extracellular space. Physiological Reviews.

2008;88(4):1277-1340.

19. Cheer JF, Aragona BJ, Heien ML, Seipel AT, Carelli RM, Wightman RM. Coordinated

accumbal dopamine release and neural activity drive goal-directed behavior. Neuron.

2007;54(2):237-244.

20. Kiyatkin EA, Brown PL. I.v. cocaine induces rapid, transient excitation of striatal

neurons via its action on peripheral neural elements: single-cell, iontophoretic study in

awake and anesthetized rats. Neuroscience. 2007;148(4):978-995. Epub 2007 Jul 2017.

21. Lerner EN, van Zanten EH, Stewart GR. Enhanced delivery of octreotide to the brain via

transnasal iontophoretic administration. Journal of Drug Targeting. 2004;12(5):273-280.

22. Probstein RF. Physicochemical Hydrodynamics, An Introduction: Second Edition; 1994.

23. Delgado AV, Gonzalez-Caballero F, Hunter RJ, Koopal LK, Lyklema J. Measurement

and interpretation of electrokinetic phenomena. Journal of Colloid and Interface Science.

2007;309(2):194-224. Epub 2007 Mar 2021.

24. Miyake M, Kurihara K. Resting potential of the mouse neuroblastoma cells. II.

Significant contribution of the surface potential to the resting potential of the cells under

physiological conditions. Biochimica et Biophysica Acta. 1983;762(2):256-264.

125 25. Viskari PJ, Landers JP. Unconventional detection methods for microfluidic devices.

Electrophoresis. 2006;27(9):1797-1810.

26. Brewer GJ. Exocytosis and directed movement of cytoplasmic vesicles toward the plasma

membrane. Journal of Theoretical Biology. 1980;85(1):75-82.

27. Stoilova S, Kancheva M. Microelectrophoretic properties of rat brain membrane

structures - synaptosomes. Studia Biophysica. 1986;113(1-2):27-30.

28. Chammas P, Federspiel WJ, Eisenberg SR. A Microcontinuum Model of Electrokinetic

Coupling in the Extracellular Matrix: Perturbation Formulation and Solution. Journal of

Colloid and Interface Science. 1994;168(2):526-538.

29. Bath BD, Lee RD, White HS, Scott ER. Imaging Molecular Transport in Porous

Membranes. Observation and Analysis of Electroosmotic Flow in Individual Pores Using

the Scanning Electrochemical Microscope. Analytical Chemistry. 1998;70(6):1047-1058.

30. Uitto OD, White HS. Electroosmotic pore transport in human skin. Pharmaceutical

Research. 2003;20(4):646-652.

31. Ranck JB, Jr. Synaptic "Learning" Due to Electroosmosis: a Theory. Science.

1964;144:187-189.

32. Anderova M, Kubinova S, Mazel T, et al. Effect of elevated K(+), hypotonic stress, and

cortical spreading depression on astrocyte swelling in GFAP-deficient mice. Glia.

2001;35(3):189-203.

33. Cragg SJ, Nicholson C, Kume-Kick J, Tao L, Rice ME. Dopamine-mediated volume

transmission in midbrain is regulated by distinct extracellular geometry and uptake.

Journal of Neurophysiology. 2001;85(4):1761-1771.

126 34. Cserr HF, DePasquale M, Nicholson C, Patlak CS, Pettigrew KD, Rice ME. Extracellular

volume decreases while cell volume is maintained by ion uptake in rat brain during acute

hypernatremia. Journal of Physiology. 1991;442:277-295.

35. Homola A, Zoremba N, Slais K, Kuhlen R, Sykova E. Changes in diffusion parameters,

energy-related metabolites and glutamate in the rat cortex after transient

hypoxia/ischemia. Neuroscience Letters. 2006;404(1-2):137-142. Epub 2006 Jun 2006.

36. Hrabetova S. Extracellular diffusion is fast and isotropic in the stratum radiatum of

hippocampal CA1 region in rat brain slices. Hippocampus. 2005;15(4):441-450.

37. Hrabetova S, Nicholson C. Dextran decreases extracellular tortuosity in thick-slice

ischemia model. Journal of Cerebral Blood Flow Metabolism. 2000;20(9):1306-1310.

38. Krizaj D, Rice ME, Wardle RA, Nicholson C. Water compartmentalization and

extracellular tortuosity after osmotic changes in cerebellum of Trachemys scripta.

Journal of Physiology. 1996;492(Pt 3):887-896.

39. Kume-Kick J, Mazel T, Vorisek I, Hrabetova S, Tao L, Nicholson C. Independence of

extracellular tortuosity and volume fraction during osmotic challenge in rat neocortex.

Journal of Physiology. 2002;542(Pt 2):515-527.

40. Lehmenkuhler A, Sykova E, Svoboda J, Zilles K, Nicholson C. Extracellular space

parameters in the rat neocortex and subcortical white matter during postnatal

development determined by diffusion analysis. Neuroscience. 1993;55(2):339-351.

41. Mazel T, Richter F, Vargova L, Sykova E. Changes in extracellular space volume and

geometry induced by cortical spreading depression in immature and adult rats.

Physiological Research. 2002;51(Suppl 1):S85-93.

127 42. Mazel T, Simonova Z, Sykova E. Diffusion heterogeneity and anisotropy in rat

hippocampus. Neuroreport. 1998;9(7):1299-1304.

43. McBain CJ, Traynelis SF, Dingledine R. Regional variation of extracellular space in the

hippocampus. Science. 1990;249(4969):674-677.

44. Nicholson C, Phillips JM. Ion diffusion modified by tortuosity and volume fraction in the

extracellular microenvironment of the rat cerebellum. Journal of Physiology.

1981;321:225-257.

45. Perez-Pinzon MA, Tao L, Nicholson C. Extracellular potassium, volume fraction, and

tortuosity in rat hippocampal CA1, CA3, and cortical slices during ischemia. Journal of

Neurophysiology. 1995;74(2):565-573.

46. Piet R, Vargova L, Sykova E, Poulain DA, Oliet SH. Physiological contribution of the

astrocytic environment of neurons to intersynaptic crosstalk. Proceedings of the National

Academy of Sciences of the United States of America. 2004;101(7):2151-2155. Epub

2004 Feb 2156.

47. Rice ME, Okada YC, Nicholson C. Anisotropic and heterogeneous diffusion in the turtle

cerebellum: implications for volume transmission. Journal of Neurophysiology.

1993;70(5):2035-2044.

48. Slais K, Vorisek I, Zoremba N, Homola A, Dmytrenko L, Sykova E. Brain metabolism

and diffusion in the rat cerebral cortex during pilocarpine-induced status epilepticus.

Experimental Neurology. 2008;209(1):145-154. Epub 2007 Sep 2020.

49. Sykova E, Fiala J, Antonova T, Vorisek I. Extracellular space volume changes and

diffusion barriers in rats with kaolin-induced and inherited hydrocephalus. European

Journal of Pediatric Surgery. 2001;11(Suppl 1):S34-37.

128 50. Sykova E, Mazel T, Hasenohrl RU, et al. Learning deficits in aged rats related to

decrease in extracellular volume and loss of diffusion anisotropy in hippocampus.

Hippocampus. 2002;12(2):269-279.

51. Sykova E, Mazel T, Simonova Z. Diffusion constraints and neuron-glia interaction during

aging. Experimental Gerontology. 1998;33(7-8):837-851.

52. Sykova E, Vorisek I, Antonova T, et al. Changes in extracellular space size and geometry

in APP23 transgenic mice: a model of Alzheimer's disease. Proceedings of the National

Academy of Sciences of the United States of America. 2005;102(2):479-484. Epub 2005

Jan 2003.

53. Sykova E, Vorisek I, Mazel T, Antonova T, Schachner M. Reduced extracellular space in

the brain of tenascin-R- and HNK-1-sulphotransferase deficient mice. European Journal

of Neuroscience. 2005;22(8):1873-1880.

54. van der Toorn A, Sykova E, Dijkhuizen RM, et al. Dynamic changes in water ADC,

energy metabolism, extracellular space volume, and tortuosity in neonatal rat brain

during global ischemia. Magnetic Resonance in Medicine. 1996;36(1):52-60.

55. Vargova L, Homola A, Zamecnik J, Tichy M, Benes V, Sykova E. Diffusion parameters

of the extracellular space in human gliomas. Glia. 2003;42(1):77-88.

56. Vorisek I, Sykova E. Ischemia-induced changes in the extracellular space diffusion

parameters, K+, and pH in the developing rat cortex and corpus callosum. Journal of

Cerebral Blood Flow Metabolism. 1997;17(2):191-203.

57. Vorisek I, Sykova E. Evolution of anisotropic diffusion in the developing rat corpus

callosum. Journal of Neurophysiology. 1997;78(2):912-919.

129 58. Zoremba N, Homola A, Rossaint R, Sykova E. Brain metabolism and extracellular space

diffusion parameters during and after transient global hypoxia in the rat cortex.

Experimental Neurology. 2007;203(1):34-41. Epub 2006 Sep 2007.

59. Nicholson C, Tao L. Hindered diffusion of high molecular weight compounds in brain

extracellular microenvironment measured with integrative optical imaging. Biophysical

Journal. 1993;65(6):2277-2290.

60. Tao L, Nicholson C. Diffusion of albumins in rat cortical slices and relevance to volume

transmission. Neuroscience. 1996;75(3):839-847.

61. Thorne RG, Nicholson C. In vivo diffusion analysis with quantum dots and dextrans

predicts the width of brain extracellular space. Proceedings of the National Academy of

Sciences of the United States of America. 2006;103(14):5567-5572. Epub 2006 Mar

5527.

62. Xiao F, Nicholson C, Hrabe J, Hrabetova S. Diffusion of flexible random-coil dextran

polymers measured in anisotropic brain extracellular space by integrative optical

imaging. Biophysical Journal. 2008;95(3):1382-1392. Epub 2008 May 1382.

63. Prokopova-Kubinova S, Vargova L, Tao L, et al. Poly[N-(2-hydroxypropyl)

methacrylamide] polymers diffuse in brain extracellular space with same tortuosity as

small molecules. Biophysical Journal. 2001;80(1):542-548.

64. Stroh M, Zipfel WR, Williams RM, Ma SC, Webb WW, Saltzman WM. Multiphoton

microscopy guides neurotrophin modification with poly(ethylene glycol) to enhance

interstitial diffusion. Nature Materials. 2004;3(7):489-494.

65. Thorne RG, Hrabetova S, Nicholson C. Diffusion measurements for drug design. Nature

Materials. 2005;4(10):713-713.

130 66. Fenstermacher J, Kaye T. Drug “Diffusion” within the Brain. Annals of the New York

Academy of Sciences. 1988;531(1):29-39.

67. Fenstermacher JD, Patlak CS, Blasberg RG. Transport of material between brain

extracellular fluid, brain cells and blood. Federation Proceedings. 1974;33(9):2070-2074.

68. Patlak CS, Fenstermacher JD. Measurements of dog blood-brain transfer constants by

ventriculocisternal perfusion. American Journal of Physiology. 1975;229(4):877-884.

69. Rosenberg GA, Kyner WT, Estrada E. Bulk flow of brain interstitial fluid under normal

and hyperosmolar conditions. The American Journal of Physiology. 1980;238(1):F42-49.

70. Kohling R, Lehmenkukhler A, Nicholson C, Speckmann EJ. Superfusion of verapamil on

the cerebral cortex does not suppress epileptic discharges due to restricted diffusion (rats,

in vivo). Brain Research. 1993;626(1-2):149-155.

71. Lehmenkuhler A, Kersting U, Nicholson C. Diffusion of penicillin in agar and cerebral

cortex of the rat. Brain Research. 1988;444(1):181-183.

72. Lucke A, Lehmenkuhler A, Altrup U, et al. Diffusion analysis of valproate and trans-2-

en-valproate in agar and in cerebral cortex of the rat. Brain Research. 1993;631(2):187-

190.

73. Nicholson C, Rice ME. Calcium diffusion in the brain cell microenvironment. Canadian

Journal of Physiology and Pharmacology. 1987;65(5):1086-1091.

74. Schwindt W, Nicholson C, Lehmenkuhler A. Critical volume of rat cortex and

extracellular threshold concentration for a pentylenetetrazol-induced epileptic focus.

Brain Research. 1997;753(1):86-97.

75. Thorne RG, Lakkaraju A, Rodriguez-Boulan E, Nicholson C. In vivo diffusion of

lactoferrin in brain extracellular space is regulated by interactions with heparan sulfate.

131 Proceedings of the National Academy of Sciences of the United States of America.

2008;105(24):8416-8421. Epub 2008 Jun 8419.

76. Barrande M, Bouchet R, Denoyel R. Tortuosity of Porous Particles. Analytical

Chemistry. 2007;79(23):9115-9121.

77. Guy Y, Muha RJ, Sandberg M, Weber SG. Determination of Zeta-Potential and

Tortuosity in Rat Organotypic Hippocampal Cultures from Electroosmotic Velocity

Measurements under Feedback Control. Analytical Chemistry. 2009;81(8):3001-3007.

78. Guy Y, Sandberg M, Weber SG. Determination of ζ-potential in rat organotypic

hippocampal cultures. Biophysical Journal. 2008;94(11):4561-4569.

79. Barrande M, Bouchet R, Denoyel R. Tortuosity of Porous Particles. Analytical Chemistry

(Washington, DC, United States). 2007;79(23):9115-9121.

80. Glover PWJ, Gomez JB, Meredith PG. Fracturing in saturated rocks undergoing triaxial

deformation using complex electrical conductivity measurements: experimental study.

Earth and Planetary Science Letters. 2000;183(1-2):201-213.

81. Shiga T. Deformation and Viscoelastic Behavior of Polymer Gels in Electric Fields.

Neutron Spin Echo Spectroscopy Viscoelasticity Rheology. Vol 134: Springer Berlin /

Heidelberg; 1997:131-163.

82. Jabbari E, Tavakoli J, Sarvestani AS. Swelling characteristics of acrylic acid

polyelectrolyte hydrogel in a dc electric field. Smart Materials and Structures.

2007;16(5):1614.

83. Gahwiler BH, Capogna M, Debanne D, McKinney RA, Thompson SM. Organotypic

slice cultures: a technique has come of age. Trends in Neurosciences. 1997;20(10):471-

477.

132 84. Stoppini L, Buchs PA, Muller D. A simple method for organotypic cultures of nervous

tissue. Journal of neuroscience methods. 1991;37(2):173-182.

85. Jaquins-Gerstl A, Michael AC. Comparison of the brain penetration injury associated

with microdialysis and voltammetry. Journal of Neuroscience Methods.

2009;183(2):127-135.

86. Mitala CM, Wang Y, Borland LM, et al. Impact of microdialysis probes on vasculature

and dopamine in the rat striatum: a combined fluorescence and voltammetric study.

Journal of Neuroscience Methods. 2008;174(2):177-185. Epub 2008 Jul 2015.

87. Budygin EA, Kilpatrick MR, Gainetdinov RR, Wightman RM. Correlation between

behavior and extracellular dopamine levels in rat striatum: comparison of microdialysis

and fast-scan cyclic voltammetry. Neuroscience Letters. 2000;281(1):9-12.

88. Kawagoe KT, Zimmerman JB, Wightman RM. Principles of voltammetry and

microelectrode surface states. Journal of Neuroscience Methods. 1993;48(3):225-240.

89. Robinson DL, Venton BJ, Heien ML, Wightman RM. Detecting subsecond dopamine

release with fast-scan cyclic voltammetry in vivo. Clinical Chemistry. 2003;49(10):1763-

1773.

90. Asai S, Kohno T, Ishii Y, Ishikawa K. A newly developed procedure for monitoring of

extracellular proteins using a push-pull microdialysis. Analytical Biochemistry.

1996;237(2):182-187.

91. Cellar NA, Burns ST, Meiners J-C, Chen H, Kennedy RT. Microfluidic Chip for Low-

Flow Push-Pull Perfusion Sampling in Vivo with On-Line Analysis of Amino Acids.

Analytical Chemistry. 2005;77(21):7067-7073.

133 92. Myers RD, Rezvani AH, Gurley-Orkin LA. New double-lumen polyethylene cannula for

push-pull perfusion of brain tissue in vivo. Journal of neuroscience methods.

1985;12(3):205-218.

93. Xu H, Guy Y, Hamsher A, Shi G, Sandberg M, Weber SG. Electroosmotic Sampling.

Application to Determination of Ectopeptidase Activity in Organotypic Hippocampal

Slice Cultures. Analytical Chemistry. 2010;82(15):6377-6383.

94. Tsien RY. A non-disruptive technique for loading calcium buffers and indicators into

cells. Nature. 1981;290(5806):527-528.

95. Cobbold PH, Rink TJ. Fluorescence and bioluminescence measurement of cytoplasmic

free calcium. Biochemistry Journal. 1987;248(2):313-328.

96. Jobsis PD, Rothstein EC, Balaban RS. Limited utility of acetoxymethyl (AM)-based

intracellular delivery systems, in vivo: interference by extracellular esterases. Journal of

Microscopy. 2007;226(Pt 1):74-81.

97. Hamsher AE, Xu H, Guy Y, Sandberg M, Weber SG. Minimizing Tissue Damage in

Electroosmotic Sampling. Analytical Chemistry. 2010;82(15):6370-6376.

98. Nicholson C, Chen KC, Hrabetova S, Tao L. Diffusion of molecules in brain extracellular

space: theory and experiment. Progress in Brain Research. 2000;125(Volume

Transmission Revisited):129-154.

99. Nicholson C, Sykova E. Extracellular space structure revealed by diffusion analysis.

Trends in Neurosciences. 1998;21(5):207-215.

100. Sykova E, Mazel T, Vargova L, Vorisek I, Prokopova-Kubinova S. Extracellular space

diffusion and pathological states. Progress in Brain Research. 2000;125:155-178.

134 101. Binder DK, Papadopoulos MC, Haggie PM, Verkman AS. In vivo measurement of brain

extracellular space diffusion by cortical surface photobleaching. Journal of Neuroscience.

2004;24(37):8049-8056.

102. Ashwal S, Holshouser BA, Tong KA. Use of Advanced Neuroimaging Techniques in the

Evaluation of Pediatric Traumatic Brain Injury. Developmental Neuroscience (Basel,

Switzerland). 2006;28(4-5):309-326.

103. Kaiser LG, Meersmann T, Logan JW, Pines A. Visualization of gas flow and diffusion in

porous media. Proceedings of the National Academy of Sciences of the United States of

America. 2000;97(6):2414-2418.

104. Kroenke CD, Neil JJ. Use of magnetic resonance to measure molecular diffusion within

the brain extracellular space. Neurochemistry International. 2004;45(4):561-568.

105. Pierpaoli C, Jezzard P, Basser PJ, Barnett A, Di Chiro G. Diffusion tensor MR imaging

of the human brain. Radiology. 1996;201(3):637-648.

106. Thorne RG, Hrabetova S, Nicholson C. Diffusion of epidermal growth factor in rat brain

extracellular space measured by integrative optical imaging. Journal of Neurophysiology.

2004;92(6):3471-3481.

107. Chen KC, Hoistad M, Kehr J, Fuxe K, Nicholson C. Theory relating in vitro and in vivo

microdialysis with one or two probes. J. Neurochem. 2002;81(1):108-121.

108. Kehr J, Hoistad M, Fuxe K. Diffusion of radiolabeled dopamine, its metabolites and

mannitol in the rat striatum studied by dual-probe microdialysis. Progress in Brain

Research. 2000;125(Volume Transmission Revisited):179-190.

135 109. Thorne RG, Nicholson C. In vivo diffusion analysis with quantum dots and dextrans

predicts the width of brain extracellular space. Proceedings of the National Academy of

Sciences of the United States of America. 2006;103(14):5567-5572.

110. Perez-Pinzon MA, Tao L, Nicholson C. Extracellular potassium, volume fraction, and

tortuosity in rat hippocampal CA1, CA3, and cortical slices during ischemia. Journal of

Neurophysiology. 1995;74(2):565-573.

111. Sykova E. Extrasynaptic volume transmission and diffusion parameters of the

extracellular space. Neuroscience (Oxford). 2004;129(4):861-876.

112. Mazel T, Simonova Z, Sykova E. Diffusion heterogeneity and anisotropy in rat

hippocampus. Neuroreport. 1998;9(7):1299-1304.

113. Rice ME. Distinct regional differences in dopamine-mediated volume transmission.

Progress in Brain Research. 2000;125(Volume Transmission Revisited):277-290.

114. Wightman RM, Amatore C, Engstrom RC, et al. Real-time characterization of dopamine

overflow and uptake in the rat striatum. Neuroscience (Oxford). 1988;25(2):513-523.

115. Strigini M. Mechanisms of morphogen movement. Journal of Neurobiology.

2005;64(4):324-333.

116. Borland LM, Shi G, Yang H, Michael AC. Voltammetric study of extracellular dopamine

near microdialysis probes acutely implanted in the striatum of the anesthetized rat.

Journal of Neuroscience Methods. 2005;146(2):149-158.

117. Savtchenko LP, Kulahin N, Korogod SM, Rusakov DA. Electric fields of synaptic

currents could influence diffusion of charged neurotransmitter molecules. Synapse (New

York). 2004;51(4):270-278.

136 118. Ardolino G, Bossi B, Barbieri S, Priori A. Non-synaptic mechanisms underlie the after-

effects of cathodal transcutaneous direct current stimulation of the human brain. Journal

of Physiology (London). 2005;568(2):653-663.

119. Talelli P, Rothwell J. Does brain stimulation after stroke have a future? Current opinion

in neurology. 2006;19(6):543-550.

120. Fregni F, Boggio Paulo S, Nitsche Michael A, Rigonatti Sergio P, Pascual-Leone A.

Cognitive effects of repeated sessions of transcranial direct current stimulation in patients

with depression. Depression and Anxiety. 2006;23(8):482-484.

121. Juckel G, Mendlin A, Jacobs BL. Electrical stimulation of rat medial prefrontal cortex

enhances forebrain serotonin output: implications for electroconvulsive therapy and

transcranial magnetic stimulation in depression. Neuropsychopharmacology.

1999;21(3):391-398.

122. Lisanby SH, Belmaker RH. Animal models of the mechanisms of action of repetitive

transcranial magnetic stimulation (RTMS): comparisons with electroconvulsive shock

(ECS). Depression and anxiety. 2000;12(3):178-187.

123. Kuroda Y, Motohashi N, Ito H, et al. Effects of repetitive transcranial magnetic

stimulation on [11C]raclopride binding and cognitive function in patients with

depression. Journal of Affective Disorders. 2006;95(1-3):35-42.

124. Kim EJ, Kim WR, Chi SE, et al. Repetitive transcranial magnetic stimulation protects

hippocampal plasticity in an animal model of depression. Neuroscience Letters.

2006;405(1-2):79-83.

137 125. Wagner T, Gangitano M, Romero R, et al. Intracranial measurement of current densities

induced by transcranial magnetic stimulation in the human brain. Neuroscience Letters.

2004;354(2):91-94.

126. Miranda P, C., Hallett M, Basser Peter J. The electric field induced in the brain by

magnetic stimulation: a 3-D finite-element analysis of the effect of tissue heterogeneity

and anisotropy. IEEE transactions on biomedical engineering. 2003;50(9):1074-1085.

127. Gluckman BJ, Nguyen H, Weinstein SL, Schiff SJ. Adaptive electric field control of

epileptic seizures. J. Neurosci. 2001;21(2):590-600.

128. LeBeau FEN, Malmierca MS, Rees A. Iontophoresis in vivo demonstrates a key role for

GABAA and glycinergic inhibition in shaping frequency response areas in the inferior

colliculus of guinea pig. Journal of Neuroscience. 2001;21(18):7303-7312.

129. Calegari F, Marzesco A-M, Kittler R, Buchholz F, Huttner WB. Tissue-specific RNA

interference in post-implantation mouse embryos using directional electroporation and

whole embryo culture. Differentiation (Malden, MA, United States). 2004;72(2-3):92-

102.

130. Cerda GA, Thomas JE, Allende ML, Karlstrom RO, Palma V. Electroporation of DNA,

RNA, and morpholinos into zebrafish embryos. Methods (San Diego, CA, United States).

2006;39(3):207-211.

131. Abla N, Naik A, Guy RH, Kalia YN. Contributions of electromigration and

electroosmosis to peptide iontophoresis across intact and impaired skin. Journal of

Controlled Release. 2005;108(2-3):319-330.

138 132. Imanidis G, Luetolf P. An extended model based on the modified Nernst-Planck equation

for describing transdermal iontophoresis of weak electrolytes. Journal of Pharmaceutical

Sciences. 2006;95(7):1434-1447.

133. Nair V, Panchagnula R. Physicochemical considerations in the iontophoretic delivery of a

small peptide: in vitro studies using arginine vasopressin as a model peptide.

Pharmacological Research. 2003;48(2):175-182.

134. Nugroho AK, Li GL, Danhof M, Bouwstra JA. Transdermal Iontophoresis of Rotigotine

Across Human Stratum Corneum in Vitro: Influence of pH and NaCl Concentration.

Pharmaceutical Research. 2004;21(5):844-850.

135. Pikal MJ. The role of electroosmotic flow in transdermal iontophoresis. Advanced Drug

Delivery Reviews. 2001;46(1-3):281-305.

136. Sieg A, Guy RH, Delgado-Charro MB. Electroosmosis in transdermal iontophoresis:

Implications for noninvasive and calibration-free glucose monitoring. Biophysical

Journal. 2004;87(5):3344-3350.

137. Vijh AK. Electrochemical treatment (ECT) of cancerous tumors: necrosis involving

hydrogen cavitation, chlorine bleaching, pH changes, electroosmosis. International

Journal of Hydrogen Energy. 2003;29(6):663-665.

138. Sekkat N, Naik A, Kalia YN, Glikfeld P, Guy RH. Reverse iontophoretic monitoring in

premature neonates: feasibility and potential. Journal of Controlled Release. 2002;81(1-

2):83-89.

139. Omasu F, Nakano Y, Ichiki T. Measurement of the electrophoretic mobility of sheep

erythrocytes using microcapillary chips. Electrophoresis. 2005;26(6):1163-1167.

139 140. Miyake M, Kurihara K. Resting potential of the mouse neuroblastoma cells. II.

Significant contribution of the surface potential to the resting potential of the cells under

physiological conditions. Biochim. Biophys. Acta. 1983;762(2):256-264.

141. Dunlap CA, Biresaw G, Jackson MA. Hydrophobic and electrostatic cell surface

properties of blastospores of the entomopathogenic fungus Paecilomyces fumosoroseus.

Colloids and Surfaces, B: Biointerfaces. 2005;46(4):261-266.

142. Guzelsu N, Regimbal RL. The origin of electrokinetic potentials in bone tissue: the

organic phase. Journal of biomechanics. 1990;23(7):661-672.

143. Badr GG, Waldman AA. Electrophoretic studies of intact nuclei isolated from

mammalian brain. II. Electrokinetic characteristics following treatment and comparison

of glial and neuronal nuclei. International Journal of Neuroscience. 1973;6(3):131-139.

144. Brewer GJ. Exocytosis and directed movement of cytoplasmic vesicles toward the plasma

membrane. Journal of Theoretical Biology. 1980;85(1):75-82.

145. Stoilova S, Kancheva M. Microelectrophoretic properties of rat brain membrane

structures - synaptosomes. Stud. Biophys. 1986;113(1-2):27-30.

146. Polevoi VV, Bilova TE, Shevtsov YI. Electroosmotic phenomena in plant tissues.

Biology Bulletin (Moscow, Russian Federation). 2003;30(2):133-139.

147. Miyamoto M, Nakahari T, Yoshida H, Imai Y. Electroosmotic flow measurements.

Journal of Membrane Science. 1989;41:377-391.

148. Nordenstrom BE. Fleischner lecture. Biokinetic impacts on structure and imaging of the

lung: the concept of biologically closed electric circuits. AJR. American journal of

roentgenology. 1985;145(3):447-467.

140 149. Rubashkin A, Iserovich P, Hernandez JA, Fischbarg J. Epithelial Fluid Transport:

Protruding Macromolecules and Space Charges Can Bring about Electro-Osmotic

Coupling at the Tight Junctions. Journal of Membrane Biology. 2006;208(3):251-263.

150. Boyack JR, Giddings JC. Theory of electrophoretic mobility in stabilized media. Archives

of Biochemistry and Biophysics. 1963;100(1):16-25.

151. Rathore AS, Wen E, Horvath C. Electroosmotic Mobility and Conductivity in Columns

for Capillary Electrochromatography. Analytical Chemistry. 1999;71(14):2633-2641.

152. Gahwiler BH, Capogna M, Debanne D, McKinney RA, Thompson SM. Organotypic

slice cultures: a technique has come of age. Trends in Neurosciences. 1997;20(10):471-

477.

153. Bahr BA. Long-term hippocampal slices: a model system for investigating synaptic

mechanisms and pathologic processes. Journal of Neurosci Research. 1995;42(3):294-

305.

154. Beeskow T, Kroner KH, Anspach FB. Nylon-based affinity membranes: impacts of

surface modification on protein adsorption. Journal of Colloid and Interface Science.

1997;196(2):278-291.

155. Masson J-F, Battaglia TM, Davidson MJ, et al. Biocompatible polymers for antibody

support on gold surfaces. Talanta. 2005;67(5):918-925.

156. Buchs PA, Stoppini L, Muller D. Structural modifications associated with synaptic

development in area CA1 of rat hippocampal organotypic cultures. Brain research.

1993;71(1):81-91.

157. Gahwiler BH. Organotypic cultures of neural tissue. Trends in Neuroscience.

1988;11(11):484-489.

141 158. Holopainen IE. Organotypic Hippocampal Slice Cultures: A Model System to Study

Basic Cellular and Molecular Mechanisms of Neuronal Cell Death, Neuroprotection, and

Synaptic Plasticity. Neurochemical Research. 2005;30(12):1521-1528.

159. Olofsson J, Levin M, Stroemberg A, Weber SG, Ryttsen F, Orwar O. Scanning

electroporation of selected areas of adherent cell cultures. Analytical Chemistry

(Washington, DC, United States). 2007;79(12):4410-4418.

160. Todd P, Plank LD, Kunze ME, et al. Electrophoretic separation and analysis of living

cells from solid tissues by several methods. Human embryonic kidney cell cultures as a

model. Journal of Chromatography. 1986;364:11-24.

161. Vornov JJ, Tasker RC, Coyle JT. Direct observation of the agonist-specific regional

vulnerability to glutamate, NMDA, and kainate neurotoxicity in organotypic

hippocampal cultures. Experimental. Neurology. 1991;114(1):11-22.

162. Prokopova-Kubinova S, Vargova L, Tao L, et al. Poly[N-(2-hydroxypropyl)

methacrylamide] polymers diffuse in brain extracellular space with same tortuosity as

small molecules. Biophysical Journal. 2001;80(1):542-548.

163. Nicholson C, Tao L. Hindered diffusion of high molecular weight compounds in brain

extracellular microenvironment measured with integrative optical imaging. Biophysical

Journal. 1993;65(6):2277-2290.

164. McBain CJ, Traynelis SF, Dingledine R. Regional variation of extracellular space in the

hippocampus. Science. 1990;249(4969):674-677.

165. Burdalo J, Medrano R, Saiz E, Tarazona MP. A simple method to determine unperturbed

dimensions of polymers using size exclusion chromatography and multiangle light

scattering. Polymer. 1999;41(4):1615-1620.

142 166. Sulzer D. The complex regulation of dopamine output: A review of current themes.

Clinical Neuroscience Research. 2005;5(2-4):117-121.

167. Watts RJ, Harrigan RW, Crosby GA. Temperature and magnetic-field effects on the

luminescence lifetimes of complexes of ruthenium(II) and iridium(III). Chemical Physics

Letters. 1971;8(1):49-50.

168. Demas JN, Harris EW, McBride RP. Energy transfer from luminescent transition metal

complexes to oxygen. Journal of the American Chemical Society. 1977;99(11):3547-

3551.

169. Zhou F, Zhu X, Castellani RJ, et al. Hibernation, a model of neuroprotection. The

American journal of pathology. 2001;158(6):2145-2151.

170. Stone TW. Microiontophoresis and pressure ejection. Vol 8. Chichester: Wiley; 1985.

171. Nicholson C, Tao L. Hindered diffusion of high molecular weight compounds in brain

extracellular microenvironment measured with integrative optical imaging. Biophysical

Journal. 1993;65(6):2277-2290.

172. Prokopova-Kubinova S, Vargova L, Tao L, et al. Poly[N-(2-hydroxypropyl)

methacrylamide] polymers diffuse in brain extracellular space with same tortuosity as

small molecules. Biophysical Journal. 2001;80(1):542-548.

173. Samec Z, Marecek V, Koryta J, Khalil MW. Investigation of ion transfer across the

interface between two immiscible electrolyte solutions by cyclic voltammetry. Journal of

Electroanalytical Chemistry and Interfacial Electrochemistry. 1977;83(2):393-397.

174. Nischang I, Tallarek U. Fluid dynamics in capillary and chip electrochromatography.

Electrophoresis. 2007;28(4):611-626.

143 175. Maiti S, Shear JB, Williams RM, Zipfel WR, Webb WW. Measuring serotonin

distribution in live cells with three-photon excitation. Science. 1997;275(5299):530-532.

176. Pellegatti L, Zhang J, Drahos B, et al. Pyridine-based lanthanide complexes: towards

bimodal agents operating as near infrared luminescent and MRI reporters. Chemical

Communications. 2008(48):6591-6593.

177. Kassam Amin B, Engh Johnathan A, Mintz Arlan H, Prevedello Daniel M. Completely

endoscopic resection of intraparenchymal brain tumors. Journal of neurosurgery.

2009;110(1):116-123.

178. Hao J, Li SK, Liu C-Y, Kao WWY. Electrically assisted delivery of macromolecules into

the corneal epithelium. Experimental Eye Research. 2009;89(6):934-941.

179. Hughes M. Ocular drug delivery: expanding the field of vision. Drug Delivery

Technology. 2009;9(4):34-35.

180. Colombo R, Brausi M, Da Pozzo L, et al. Thermo-chemotherapy and electromotive drug

administration of mitomycin C in superficial bladder cancer eradication. a pilot study on

marker lesion. European urology. 2001;39(1):95-100.

181. Clarke G, Hill RG, Simmonds MA. Microiontophoretic release of drugs from

micropipettes: use of 24Na as a model. British Journal of Pharmacology.

1973;48(1):156-161.

182. Hoffer BJ, Neff NH, Siggins GR. Microiontophoretic release of norepinephrine from

micropipettes. Neuropharmacology. 1971;10(21):175-180.

183. Bloom FE. To spritz or not to spritz: the doubtful value of aimless iontophoresis. Life

sciences. 1974;14(10):1819-1834.

144 184. Shoemaker WJ, Balentine LT, Siggins GR, Hoffer BJ, Henriksen SJ, Bloom FE.

Characteristics of the release of adenosine 3':5'-monophosphate from micropipets by

microiontophoresis. Journal of cyclic nucleotide research. 1975;1(2):97-106.

185. Purves R. Ionophoresis - progress and pitfalls. Trends in Neurosciences. 1980;3(10):245-

247.

186. Chen KC, Nicholson C. Measurement of diffusion parameters using a sinusoidal

iontophoretic source in rat cortex. Journal of Neuroscience Methods. 2002;122(1):97-

108.

187. Hrabetova S, Masri D, Tao L, Xiao F, Nicholson C. Calcium diffusion enhanced after

cleavage of negatively charged components of brain extracellular matrix by

chondroitinase ABC. Journal of Physiology. 2009;587(16):4029-4049.

188. Dionne VE. Characterization of drug iontophoresis with a fast microassay technique.

Biophysical Journal. 1976;16(7):705-717.

189. Armstrong-James M, Fox K, Kruk ZL, Millar J. Quantitative ionophoresis of

catecholamines using multibarrel carbon fibre microelectrodes. Journal of Neuroscience

Methods. 1981;4(4):385-406.

190. Armstrong-James M, Millar J, Kruk ZL. Quantification of noradrenaline iontophoresis.

Nature. 1980;288(5787):181-183.

191. Herr NR, Daniel KB, Belle AM, Carelli RM, Wightman RM. Probing presynaptic

regulation of extracellular dopamine with iontophoresis. ACS Chemical Neuroscience.

2010;1(9):627-638.

192. Purves RD. The physics of iontophoretic pipettes. Journal of neuroscience methods.

1979;1(2):165-178.

145 193. Curtis DR, Perrin DD, Watkins JC. Excitation of spinal neurons by the ionophoretic

application of agents which chelate calcium. Journal of Neurochemistry. 1960;6:1-20.

194. Trubatch J, Van Harreveld A. Spread of iontophoretically injected ions in a tissue.

Journal of Theoretical Biology. 1972;36(2):355-366.

195. Lux HD, Neher E. The equilibration time course of (K + ) 0 in cat cortex. Experimental

brain research. 1973;17(2):190-205.

196. Nicholson C, Phillips JM, Gardner-Medwin AR. Diffusion from an iontophoretic point

source in the brain: role of tortuosity and volume fraction. Brain research.

1979;169(3):580-584.

197. Norman RS. Diffusional spread of iontophoretically injected ions. Journal of theoretical

biology. 1975;52(1):159-162.

198. Herr NR, Kile BM, Carelli RM, Wightman RM. Electroosmotic flow and its contribution

to iontophoretic delivery. Analytical Chemistry. 2008;80(22):8635-8641.

199. Scott ER, Laplaza AI, White HS, Phipps JB. Transport of ionic species in skin:

contribution of pores to the overall skin conductance. Pharmaceutical research.

1993;10(12):1699-1709.

200. Drouin H. On microelectrode ionophoresis. Biophysical Journal. 1984;46(5):597-604.

201. Noraberg J, Kristensen BW, Zimmer J. Markers for neuronal degeneration in organotypic

slice cultures. Brain research protocols. 1999;3(3):278-290.

202. Beisler AT, Schaefer KE, Weber SG. Simple method for the quantitative examination of

extra column band broadening in microchromatographic systems. Journal of

Chromatography A. 2003;986(2):247-251.

146 203. Barz DPJ, Vogel MJ, Steen PH. Determination of the Zeta Potential of Porous Substrates

by Droplet Deflection. I. The Influence of Ionic Strength and pH Value of an Aqueous

Electrolyte in Contact with a Borosilicate Surface. Langmuir. 2009;25(3):1842-1850.

204. McGuffin VL, Tavares MFM. Computer-Assisted Optimization of Separations in

Capillary Zone Electrophoresis. Analytical Chemistry. 1997;69(2):152-164.

205. Thormann W, Zhang C-X, Caslavska J, Gebauer P, Mosher RA. Modeling of the impact

of ionic strength on the electroosmotic flow in capillary electrophoresis with uniform and

discontinuous buffer systems. Analytical Chemistry. 1998;70(3):549-562.

206. Scales N, Tait RN. Modeling electroosmotic and pressure-driven flows in porous

microfluidic devices: Zeta potential and porosity changes near the channel walls. Journal

of Chemical Physics. 2006;125(9):094714/094711-094714/094712.

207. Zoski CG, Bond AM, Allinson ET, Oldham KB. How long does it take a microelectrode

to reach a voltammetric steady state? Analytical Chemistry. 1990;62(1):37-45.

208. Krnjević K. When and why amino acids? The Journal of Physiology. 2010;588(1):33-44.

209. Kirby BJ, Hasselbrink EF. Zeta potential of microfluidic substrates: 1. Theory,

experimental techniques, and effects on separations. Electrophoresis. 2004;25(2):187-

202.

210. Scales N, Tait RN. Modeling wall effects in capillary electrochromatography. Journal of

Chromatography, A. 2008;1205(1-2):150-157.

211. Rathore AS, Horvath C. Axial Nonuniformities and Flow in Columns for Capillary

Electrochromatography. Analytical Chemistry. 1998;70(14):3069-3077.

147 212. Westerhuis WH, Sturgis JN, Niederman RA. Reevaluation of the electrophoretic

migration behavior of soluble globular proteins in the native and detergent-denatured

states in polyacrylamide gels. Analytical Biochemistry. 2000;284(1):143-152.

213. Stylianopoulos T, Diop-Frimpong B, Munn LL, Jain RK. Diffusion Anisotropy in

Collagen Gels and Tumors: The Effect of Fiber Network Orientation. Biophysical

Journal. 2010;99(10):3119-3128.

214. Baraban SC, Bausch SB. Organotypic Hippocampal Slice Cultures as a Model of Limbic

Epileptogenesis. Animal Models of Epilepsy. Vol 40: Humana Press; 2009:183-201.

215. Hamsher A, Xu H, Guy Y, Sandberg M, Weber SG. Minimizing tissue damage in

electroosmotic sampling. Analytical Chemistry. 2010;82(15):6370-6376.

216. Duport S, Robert F, Muller D, Grau G, Parisi L, Stoppini L. An in vitro blood-brain

barrier model: cocultures between endothelial cells and organotypic brain slice cultures.

Proceedings of the National Academy of Sciences of the United States of America.

1998;95(4):1840-1845.

217. Oldenziel WH, Dijkstra G, Cremers TIFH, Westerink BHC. Evaluation of Hydrogel-

Coated Glutamate Microsensors. Analytical Chemistry. 2006;78(10):3366-3378.

218. Kim J, Tsien RW. Synapse-specific adaptations to inactivity in hippocampal circuits

achieve homeostatic gain control while dampening network reverberation. Neuron.

2008;58(6):925-937.

219. Lipinski HG. Brain tissue slice thickness monitored by ion-profile measurement. Journal

of neuroscience methods. 1992;42(1-2):37-43.

148 220. Oldenziel WH, van der Zeyden M, Dijkstra G, et al. Monitoring extracellular glutamate

in hippocampal slices with a microsensor. Journal of Neuroscience Methods.

2007;160(1):37-44.

221. Robert F, Parisi L, Bert L, Renaud B, Stoppini L. Microdialysis monitoring of

extracellular glutamate combined with the simultaneous recording of evoked field

potentials in hippocampal organotypic slice cultures. Journal of Neuroscience Methods.

1997;74(1):65-76.

222. Schrenk-Siemens K, Perez-Alcala S, Richter J, et al. Embryonic stem cell-derived

neurons as a cellular system to study gene function: Lack of amyloid precursor proteins

APP and APLP2 leads to defective synaptic transmission. Stem Cells. 2008;26(8):2153-

2163.

223. Takahashi M, Liou SY, Kunihara M. Ca(2+)- and Cl(-)-dependent, NMDA receptor-

mediated neuronal death induced by depolarization in rat hippocampal organotypic

cultures. Brain Research. 1995;675(1-2):249-256.

224. Cho S, Wood A, Bowlby MR. Brain slices as models for neurodegenerative disease and

screening platforms to identify novel therapeutics. Current Neuropharmacology.

2007;5(1):19-33.

225. Noraberg J. Organotypic brain slice cultures: An efficient and reliable method for

neurotoxicological screening and mechanistic studies. ATLA, Alternatives to Laboratory

Animals. 2004;32(4):329-337.

226. Muller D, Toni N, Buchs P-A, Parisi L, Stoppini L. Interface Organotypic Hippocampal

Slice Cultures. Protocols for Neural Cell Culture: Springer Protocols; 2001:13-27.

149